Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Does fat provide energy for breast tumour cell invasion and metastasis? Morgan Jones Supervisors Dr. Margaret Currie Dr. Elisabeth Phillips Prof. Mark Hampton A thesis submitted for the degree of Bachelor of Biomedical Science with Honours University of Otago, Christchurch 2015 i Abstract Breast cancer is the most commonly diagnosed cancer in New Zealand women. Obese breast cancer patients are more likely to have tumours with advanced clinical stage and high vascular and lymph node involvement. The tumour microenvironment provides vital support for tumours during development and progression of cancer, yet the local effects of stromal adipocytes on breast cancer cells have been largely overlooked. Recent studies by Dirat et al. (2011) and Bochet et al. (2011) have shown that breast cancer cells co-cultured with adipocytes become more resistant to radio- and chemotherapy, and more invasive. However, little is known about the metabolic changes that occur in breast cancer cells when they are cultured with adipocytes. Nieman et al. (2011) determined that lipids were transferred from omental adipocytes to ovarian cancer cells and were consequently used in β-oxidation. It was hypothesised that β-oxidation is increased in breast cancer to exploit the glycerol and fatty acids released by lipolysis from adipocytes in order to support the migration and invasion of breast cancer cells. In this study, breast cancer cell lines MCF7 (ER+; oestrogen receptor positive) and MDA-MB231 (ER-/PR-/HER2-; oestrogen, progesterone and human epidermal growth factor receptor negative) were co-cultured with adipocytes isolated from breast adipose tissue. Adipose tissue samples were collected via the Cancer Society Tissue Bank from patients at Christchurch Hospital undergoing surgery for therapeutic mastectomy, prophylactic mastectomy and breast reductions. A Seahorse XF24 Analyser was used to measure oxygen consumption and extracellular acidification, as indicators of oxidative phosphorylation and glycolysis, respectively, in breast cancer cells grown alone or in co-culture with human breast adipocytes. MCF7 cells were found to have upregulated glycolysis after co-culture with adipocytes. Western blotting was used to assess differences in the expression of proteins involved in β-oxidation between breast cancer cells grown alone or in co-culture with adipocytes. Levels of carnitine palmitoyltransferase 1 (CPT1A), a protein involved in translocation of fatty acids into the mitochondrial matrix for β-oxidation, showed no change. However, phosphorylated acetyl-CoA carboxylase (ACC), a key metabolic enzyme that when inhibited relieves inhibition of CPT1A to allow fatty acid translocation into mitochondria, showed increased levels in both MCF7 and MDA-MB-231 cells after co-culture with adipocytes. These results support the concept that breast cancer cell metabolism, specifically glycolysis and β-oxidation, is being altered in the presence of adipocytes to utilise fatty acids and glycerol released by adipocytes during lipolysis. ii Acknowledgements I would like to begin by thanking my primary supervisor, Dr. Margaret Currie, for all her efforts in supporting me this year. Thank you so much for accepting me for this project and for guiding me through it. You patiently coped with my Ultimate Frisbee absences, whilst putting a little bit of “fear” in me. Time stamped emails could only hint at how many hours you put in behind the scenes to help me, despite it being a busy year for you too. A huge thank you is also due to my supervisor, Dr. Elisabeth Phillips. Heeeey Elisabeth…? Did you know you answered my endless questions, even when you hadn’t had a morning coffee yet? You’ve taught me so many valuable lab skills and I’m so happy to have had a supervisor who showed me how to celebrate with an enthusiastic data dance. Thanks as well to my last supervisor, Dr. Mark Hampton, for juggling me with his other students. Thank you for providing expertise and access to the Seahorse and for questioning me about experimental design. A special thank you to Dr. Karina O’Connor and Dr. Andree Pearson for helping me with my Seahorse troubles whenever I knocked on their doors. Thanks also to Karina for helping me with conducting experiments and interpreting the Seahorse data. I gratefully acknowledge Helen and the Cancer Society Tissue Bank for providing the adipose samples needed for this project. All the other people in the McKenzie Cancer Research Group who helped me through this, whether with lab advice or just some food and a laugh, thank you so much. To the other Honours students, we made it! Lastly, the people who didn’t have to deal with me in the lab, but had to put up with me the rest of the time. Thank you to my flatmates for putting up with my general complaints and irritability, to the Ultimate Frisbee community for allowing me to let off steam playing and partying with them, to my parents for their support in financing me and organising various parts of my life, and lastly to Ben for always being there for me and surviving all my misguided frustrations. iii List of Abbreviations Abbreviation Description ACC acetyl-CoA carboxylase AMP adenosine monophosphate AMPk AMP-activated protein kinase ATP adenosine triphosphate BMI body mass index CAA cancer associated adipocyte CAF cancer associated fibroblast CoA coenzyme A CPT1 carnitine palmitoyl transferase 1 DMEM Dulbecco’s modified eagle medium DTT dithiothreitol ER oestrogen receptor FADH2 flavin adenine dinucleotide FAO fatty acid oxidation FASN fatty acid synthase FBS foetal bovine serum FCCP carbonyl cyanide-p-trifluoromethoxyphenylhydrazone GPD2 glycerol-3-phosphate dehydrogenase HBSS Hank’s balanced salt solution HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid HER2 human epidermal growth factor receptor 2 IBMX 3-isobutyl-1-methylxanthine IL-6 interleukin 6 mGPDH mitochondrial glycerol-3-phosphate dehydrogenase MMP matrix metalloproteinase MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide NADH nicotinamide adenine dinucleotide NADPH nicotinamide adenine dinucleotide phosphate OXPHOS oxidative phosphorylation PAI-1 plasminogen activator inhibitor-1 phospho-ACC phosphorylated ACC iv PR progesterone receptor ROS reactive oxygen species RPMI Roswell Park Memorial Institute media SDS sodium dodecyl sulfate TBS tris buffered saline TBS-T tris buffered saline with tween TCA tricarboxylic acid TGF-β transforming growth factor β T3 3,3′,5-triiodo-l-thyronine VEGF vascular endothelial growth factor v Contents Abstract.............................................................................................................................ii Acknowledgements ........................................................................................................ iii List of Abbreviations .......................................................................................................iv List of Figures................................................................................................................... 5 List of Tables .................................................................................................................... 7 1 Introduction ................................................................................................................... 8 1.1 Obesity and cancer .................................................................................................. 9 1.2 Breast cancer ........................................................................................................... 9 1.2.1 Breast cancer and obesity ............................................................................... 10 1.3 Tumour microenvironment ................................................................................... 11 1.3.1 Involvement in breast cancer ......................................................................... 12 1.3.2 Cancer-Associated Adipocytes (CAAs) ......................................................... 13 1.4 Metabolism in cancer ............................................................................................ 14 1.4.1 Warburg theory .............................................................................................. 14 1.4.2 Two-compartment tumour metabolism .......................................................... 15 1.4.3 β-oxidation ..................................................................................................... 16 1.4.4 Proteins involved in β-oxidation .................................................................... 18 1.5 Summary and hypothesis ...................................................................................... 20 1.6 Aims and objectives .............................................................................................. 21 2 Materials and Methods ................................................................................................ 23 2.1 Breast adipose tissue samples and processing ...................................................... 23 2.1.1 Collection of patient samples ......................................................................... 23 2.1.2 Isolation of pre-adipocytes ............................................................................. 23 2.1.3 Culture of pre-adipocytes ............................................................................... 24 2.1.4 Differentiation of pre-adipocytes into mature adipocytes .............................. 24 2.2 Breast cancer cells ................................................................................................ 24 1 2.2.1 Passage ........................................................................................................... 25 2.3 Co-culture of mature adipocytes and breast cancer cell lines ............................... 25 2.3.1 Day 1 .............................................................................................................. 25 2.3.2 Day 4 .............................................................................................................. 26 2.3.3 Day 5 .............................................................................................................. 26 2.4 Seahorse XF24 Extracellular Flux Analyser ........................................................ 26 2.4.1 Optimisation for XF24 ................................................................................... 27 2.4.2 Mitochondrial stress test ................................................................................ 27 2.4.3 Normalisation of XF data ............................................................................... 28 2.4.4 Fatty acid oxidation test optimisation ............................................................ 29 2.5 Western Blot ......................................................................................................... 31 2.5.1 Cell lysis ......................................................................................................... 31 2.5.2 Protein quantification ..................................................................................... 31 2.5.3 SDS-PAGE ..................................................................................................... 31 2.5.4 Transfer .......................................................................................................... 32 2.5.5 Antibodies ...................................................................................................... 32 2.5.6 Detection ........................................................................................................ 33 2.5.7 Re-probing...................................................................................................... 33 2.6 Glycerol assay ....................................................................................................... 34 2.6.1 Sample collection ........................................................................................... 34 2.6.2 Detection ........................................................................................................ 34 2.7 Statistical analysis ................................................................................................. 34 2.7.1 Seahorse XF analyser ..................................................................................... 34 2.7.2 Western blotting ............................................................................................. 35 2.7.3 Glycerol Assay ............................................................................................... 35 3 Results ......................................................................................................................... 36 2 3.1 Analysis of metabolism in breast cancer cells and breast cancer cells co-cultured with adipocytes using the Seahorse XF24 Extracellular Flux Analyser ..................... 36 3.1.1 Optimisation for MCF7 and MDA-MB-231 cell lines on the Seahorse XF24 Extracellular Flux Analyser .................................................................................... 36 3.1.2 Metabolic characterisation of MCF7 and MDA-MB-231 breast cancer cells41 3.1.5 Metabolic characterisation of MCF7 breast cancer cells grown alone and in adipocyte co-culture ................................................................................................ 48 3.1.6 Fatty acid oxidation stress test optimisation .................................................. 52 3.2 Analysis of proteins involved in metabolism using Western blotting .................. 55 3.2.1 CPT1A............................................................................................................ 55 2.3 ACC and pACC................................................................................................. 57 3.2.4 GPD2 .............................................................................................................. 60 3.3 Analysis of glycerol in conditioned media ........................................................... 62 4 Discussion.................................................................................................................... 63 4.1 Analysis of metabolism in breast cancer cells and breast cancer cells co-cultured with adipocytes using the Seahorse Extracellular Flux Analyser ............................... 63 4.2 Analysis of proteins involved in β-oxidation using Western blotting .................. 69 4.3 Analysis of glycerol in conditioned media ........................................................... 71 4.4 Future Work .......................................................................................................... 73 4.5 Conclusion ............................................................................................................ 75 5 Appendices .................................................................................................................. 76 5.1 Media .................................................................................................................... 76 5.1.1 Pre-adipocyte growth media .......................................................................... 76 5.1.2 Serum free media ........................................................................................... 76 5.1.3 Adipocyte differentiation media .................................................................... 76 5.1.4 Adipocyte maintenance media ....................................................................... 76 5.1.5 Breast cancer cell media................................................................................. 77 5.1.6 MTT assay media ........................................................................................... 77 3 5.1.7 Mitochondrial stress test assay media ............................................................ 77 5.1.8 Substrate limited media .................................................................................. 77 5.1.9 FAO assay media (1X) ................................................................................... 77 5.1.10 FAO assay media (5X) ................................................................................. 78 5.2 Buffers and solutions ............................................................................................ 78 5.2.1 Phosphate buffered saline (PBS) .................................................................... 78 5.2.2 Cell-Tak solution ............................................................................................ 78 5.2.3 Solubilisation solution .................................................................................... 78 5.2.4 Lysis buffer .................................................................................................... 78 5.2.5 RIPA buffer .................................................................................................... 78 5.2.6 Sample loading buffer .................................................................................... 79 5.2.7 Running buffer ............................................................................................... 79 5.2.8 Transfer buffer ............................................................................................... 79 5.2.9 TBS-T ............................................................................................................. 79 5.2.10 Mild stripping buffer .................................................................................... 79 5.3 Supplementary Figures ......................................................................................... 80 References ...................................................................................................................... 86 4 List of Figures Figure 1.1: Interactions between tumour cells and mature adipocytes as described in text……………………………………………………………………………………..14 Figure 1.2: Key proteins and reactions in the fatty acid oxidation pathway as described in the text……………………………………………………………………………….17 Figure 1.3: Summary of metabolic changes that occur in interacting ovarian cancer cells and adipocytes as described in the text…………………………………………………18 Figure 1.4: Hypothesis of breast cancer-adipocyte interactions………………………..21 Figure 3.1: Comparison of MCF7 and MDA-MB-231 cell seeding using representative 10X photographs……………………………………………………………………….38 Figure 3.2: Pictorial representation of the changes in the electron transport chain due to oligomycin, FCCP and antimycin-A…………………………………………………...39 Figure 3.3: FCCP optimisation of MCF7 and MDA-MB-231 breast cancer cell lines…40 Figure 3.4: Representation of a typical mitochondrial stress test, identifying the fractions attributed to various metabolic parameters……………………………………………..41 Figure 3.5: Comparing oxidative metabolic parameters of MCF7 and MDA-MB-231 breast cancer cell lines………………………………………………………………….44 Figure 3.6: Energy graph displaying cells in their baseline and stressed phenotypes…..45 Figure 3.7: Comparing energy phenotypes of MCF7 and MDA-MB-231 breast cancer cell lines using OCR and ECAR in baseline and stressed conditions………………….47 Figure 3.8: Comparing oxidative metabolic parameters of MCF7 and MCF7 co-cultured with adipocytes…………………………………………………………………………49 Figure 3.9: Comparing energy phenotypes of MCF7 and adipocyte co-cultured MCF7 using OCR and ECAR in baseline and stressed conditions……………………………..51 Figure 3.10: Representation of a typical fatty acid oxidation stress test, identifying the fractions attributed to exogenous or endogenous oxidation…………………………….52 5 Figure 3.11: FCCP optimisation of MCF7 cells in the presence of BSA……………….54 Figure 3.12: MCF7 OCR graph during FCCP optimisation with BSA or BSA:palmitate………………………………………………………………………….54 Figure 3.13: Comparison of CPT1A protein level in MCF7 and MDA-MB-231 cells compared to their adipocyte co-cultured counterparts……………………………….....56 Figure 3.14: Comparison of ACC1 protein level in MCF7 and MDA-MB-231 cells compared to their adipocyte co-cultured counterparts…………………………………58 Figure 3.15: Comparison of phospho-ACC protein level in MCF7 and MDA-MB-231 cells compared to their adipocyte co-cultured counterparts……………………………59 Figure 3.16: Comparison of GPD2 protein level between MCF7 and MDA-MB-231…60 Figure 3.17: Comparison of GPD2 protein level in MCF7 and MDA-MB-231 cells compared to their adipocyte co-cultured counterparts….……………………….……...61 Figure 3.18: Comparison of glycerol levels in various conditioned media samples from MDA-MB-231 and MCF7 adipocyte co-cultures………………………………………62 Figure 5.1: Representative photographs of adipocyte samples before and after co-culture with breast cancer cells…………………………………………………………………80 Figure 5.2: Representative photographs of cell seeding for MCF7 and MDA-MB-231 from 5x105-1x104 in an XF 24-well plate………………………………………………81 Figure 5.3: Representative photographs of cell seeding for MCF7 and MDA-MB-231 from 5x105-1x104 in an XF 24-well plate………………………………………………82 Figure 5.4: MTT assay showing viability of MCF7 and MDA-MB-231 cells at varying densities………………………………………………………………………………..83 Figure 5.5: Glycerol levels in varying stages of individual adipocyte co-cultures with MCF7 and MDA-MB-231 cells………………………………………………………..84 Figure 5.6: MTT assays of MCF7 and MDA-MB-231 cells grown on inserts or not on inserts as controls to adipocyte co-cultured cells……………………………………….85 6 List of Tables Table 2.1: Mitochondrial stress test protocol…………………………………………...28 Table 2.2: Fatty acid oxidation stress test protocol……………………………………..30 Table 2.3: Primary and secondary antibody concentrations used in Western blotting….33 Table 3.1: Calculations to determine metabolic parameters of the mitochondrial stress test……………………………………………………………………………………...42 Table 3.2: Calculations to determine metabolic parameters of the fatty acid oxidation stress test……………………………………………………………………………….53 7 1 Introduction Obesity is a significant problem worldwide that is only now being fully defined. Over the last 35 years worldwide obesity has more than doubled (1). Obesity is not only a concern for those directly affected; it also impacts greatly on countries’ health systems and is therefore of economic concern (2, 3). New Zealand is one of many countries struggling with the problem of rising obesity rates; 35% of adults are overweight and an additional 31% of adults are obese (4, 5), which is a significant increase from the 19% obesity rate in 1997 (4). Obesity rates are also much greater in Pacific (68% obese) and Maori (48% obese) ethnicities, which is of specific concern to New Zealand and our health system (5). Breast cancer is the most commonly diagnosed cancer and the leading cause of cancer death in women worldwide (6). New Zealand statistics reflect this with breast cancer as the most commonly diagnosed cancer in women, accounting for 28.7% of female cancers (7). Breast cancer is also the most expensive cancer to treat using 15% of the $511 million spent every year on cancer in New Zealand (8). Obesity is known to increase the risk of post-menopausal breast cancer and contributes to poorer prognosis for all breast cancers (9, 10). The tumour microenvironment has an important role in cancer development and progression (11, 12) and is one of the ways in which obesity can have a local effect on breast cancer (13, 14). This is through breast cancer cell interactions with adipocytes, causing formation of cancer-associated adipocytes (CAA) (15). CAA can make breast cancer more aggressive, invasive, and resistant to radiotherapy and chemotherapy (15-17). One of the mechanisms CAAs may work through is by altering breast cancer cell metabolism. Stromal cells from the tumour microenvironment have been observed to provide high energy nutrients to breast cancer cells (18). It is proposed that these nutrients fuel rapid growth, proliferation, invasion and metastasis of breast cancer (19). In ovarian cancer, adipocytes have been observed to transfer lipids to neighbouring ovarian cancer cells where they are metabolised by βoxidation (17). Available fatty acids and glycerol from CAAs may be utilised by glycolysis and β-oxidation in breast cancer cells to fuel tumour growth, invasion, and metastasis. 8 1.1 Obesity and cancer Although obesity is very heterogeneous it can be broadly defined as having an excess of body fat (20). It is most commonly measured by body mass index (BMI), a calculation based on kg/m2 (21). Those with a BMI of 25-29.9 are considered overweight and those with a BMI >30 are considered obese (21). Obesity is a dangerous condition made all the more serious by its preventability. Obesity is associated with many diseases such as type 2 diabetes, cardiovascular disease, stroke, and cancer (2, 22, 23). It has been found that severely obese people die 8-10 years sooner than their normal weight counterparts (2, 23, 24). Obese people are more likely to have morbidity from chronic disorders, a poorer quality of life, and die prematurely (2, 24). Due to this acknowledged danger, there are many public health efforts focused on obesity (23, 25). Previous studies show obese and overweight people are at increased risk and have a poorer prognosis for many cancers including kidney, postmenopausal breast, oesophageal adenocarcinoma, endometrium, pancreas, gallbladder, colon, thyroid, and renal cancer (26, 27). It is estimated that being either overweight or obese contributes to 20% of all cancer diagnoses (28). Obesity is a modifiable cause of cancer and thus a useful area for research as it can be improved. Obesity has a clear link to cancer risk and progression but the mechanisms are not widely understood. It is likely that different cancers are caused through different mechanistic pathways as cancers are so variable in their aetiology (28). There is a very complex interaction between different systemic and local effects of obesity that leads to the development and progression of cancer (29, 30). The main mechanisms proposed are insulin resistance, changes in adipokine levels, inflammatory processes, and changes in obesity-related hormone levels (29, 31-33). This study will focus on the local effects of obesity on cancer; more specifically the local effects of stromal adipocytes on cancer cell metabolism. 1.2 Breast cancer Breast cancer is the abnormal development of breast cells into a malignant tumour which can result in metastasis to other locations in the body (34). This is due to mutations in genes and as such, breast cancer has been classified into at least 5 molecular subtypes: luminal A, luminal B, HER-2-enriched, claudin-low, triple negative/basal-like (35). 9 These have varying characteristics at the molecular level and consequently at a clinical level (35). The variety of breast cancer is echoed by the range of risk factors including age, age at menarche, age at pregnancy, parity, breast-feeding, abortion, exogenous hormones, as well as genetic factors (36). 1.2.1 Breast cancer and obesity As discussed above, obesity can alter the risk and prognosis of various cancers. Breast cancer is no exception. The first study looking at the relationship between breast cancer and obesity was conducted by Abe in 1976 (37, 38). They found that obese patients have poorer 5-year survival rates and are more likely to have tumours with vascular involvement, lymph node metastasis and advanced clinical stage when compared to non-obese counterparts (37). This is in line with recent findings that obese women have more distant metastases at diagnosis and a higher mortality rate (39). This poor prognosis associated with obesity has been shown to be independent of menopausal status, tumour stage and breast tumour hormone receptor status (10, 38, 40). Obesity also leads to increased risk of developing breast cancer (39, 41-43). Studies regarding obesity effects on post-menopausal and pre-menopausal breast cancer have been more unclear with most studies showing an increase in risk for post-menopausal women but a less clear link between obesity and risk in pre-menopausal women (9, 38). The reasons for higher risk and poorer prognosis of breast cancer in obese patients are unclear. The two main arguments for poor prognosis debate whether it is the biology of the breast cancer itself or the treatment of obese patients (39, 42). Obese patients show more advanced tumours at diagnosis which may be due to delayed detection (10, 42). They also have a greater likelihood of treatment failure which is probably linked to reduced chemotherapy doses in obese patients (42). However, there is also a strong argument that the tumours are not just aggressive due to detection and treatment errors but rather due to the impacts of obesity on tumour biology (10, 39, 42). 1.2.1.1 Mechanisms of obesity and breast cancer association Proposed mechanisms of obesity on breast cancer range from systemic effects, including elevated levels of circulating oestrogens, hyperinsulinemia and insulin resistance, and 10 chronic inflammation; to local effects of adipocytes in the tumour stroma, such as adipokine secretions (9, 10, 38, 43, 44). It has been found that post-menopausal obese patients have higher levels of circulating oestrogens than their normal weight counterparts (9). Oestrogen is a key hormone in normal breast development and growth and as such is involved in cancer development and progression (44). This obesity related increase in oestrogen has been associated with increased risk of breast cancer (43, 44). Obese patients also have high levels of insulin known as hyperinsulinemia which can result from insulin resistance (43). There is an increased risk of breast cancer and poorer prognosis for patients with hyperinsulinemia (43). Chronic inflammation is seen in obese patients and data suggests that this state of inflammation may influence breast cancer risk and prognosis (9, 40). These are some of the systemic mechanisms that obesity is thought to act through. One of the local mechanisms that obesity works through to alter breast cancer is adipocyte secretions, such as adipokines. Two common adipokines linked to obesity and breast cancer are leptin and adiponectin; these hormones regulate insulin sensitivity (38, 44). Leptin levels are increased in obesity (38) and adiponectin levels are decreased (44). These trends are also seen in breast cancer (38, 44, 45). Their effects may be mediated by the up regulation of aromatase which leads to higher local levels of oestrogen (44). This shows the local effect adipocytes can have on breast cancer. 1.3 Tumour microenvironment Tumours primarily consist of cancer cells with the surrounding tumour microenvironment consisting of stromal cells, immune cells, vasculature and extracellular matrix (46, 47). The microenvironment provides support for tumours and as such has a key role in the development and progression of cancer (11, 12, 48). The importance of the tumour microenvironment was noted in very early research by Paget’s “seed and soil” hypothesis (49). Paget states that for the seed to grow the soil must be hospitable; which in context means the cancer needs a nourishing environment (49, 50). This was originally hypothesised to explain why cancer is more likely to metastasise in certain organs; some organs provide a better environment or “soil” for cancer progression (49). The “seed and soil” hypothesis has recently been extended to cancer cells and the importance of their microenvironment in their development (50). 11 It is thought that the tumour and its microenvironment undergo “dynamic reciprocity”, a term developed by Mina Bissell to describe the interaction of the extracellular matrix with breast cancer cells (51, 52). This has been expanded to include the various microenvironment and tumour components and now describes how they continuously respond to each other and shape the shared environment (53). It is often debated whether the microenvironment undergoes changes which allow for initiation of cancerous growth or whether the cancer begins independently of the microenvironment and then influences it to become a support system (53, 54). It is likely that both occur and once initiated the reciprocal communication leads to the involvement of both the microenvironment and the tumour in cancer progression (53). It is thought that obesity can alter the microenvironment to make it more supportive of tumour growth (13, 14, 31). Adipose tissue in the microenvironment functions as an endocrine organ and can have systemic and local effects (55). This may be through hormonal, insulin-related, or inflammatory mechanisms (14, 55). A key area emerging in the role of obesity is the local effects of adipocytes on metabolism both in the tumour itself and the tumour microenvironment. 1.3.1 Involvement in breast cancer The specific microenvironment of breast tissue is necessary for normal mammary gland development and undergoes many changes that make it suitable for tumour development (54). Thus an altered microenvironment can contribute to breast cancer proliferation, survival and invasion (47, 56). It has been shown that breast cancer cells can lose their malignant phenotype when they are grown in a normal mammary microenvironment (57). This is one of the most convincing arguments that breast cancer requires the altered tumour microenvironment for growth and progression (47). Various cells in the microenvironment have a role in breast cancer. Genetic and epigenetic alterations occur in the cells that compose the tumour microenvironment and this is one of the many ways they can influence the tumour (58, 59). Tumour associated macrophages have a role in angiogenesis and tumour progression (53, 60) and as such are correlated with poorer prognosis (61). Mesenchymal stem cells can enhance invasion and metastasis of breast cancer (62). 12 One of the well-studied stromal players in breast cancer is the fibroblast. In the presence of breast cancer, fibroblasts undergo changes to become cancer-associated fibroblasts (CAFs) (63). CAFs are the most prominent non-cancerous cells in the tumour stroma and secrete various factors, such as VEGF (vascular endothelial growth factor), IL-6 (interleukin-6), TGF-β (transforming growth factor β), and matrix metalloproteinases (MMP) that affect nearby cancer cells (61). In doing so CAFs promote breast cancer development, angiogenesis, invasion and metastasis (61, 64-66). 1.3.2 Cancer-Associated Adipocytes (CAAs) Adipocytes (fat cells) are often overlooked as tumour stromal cells and disregarded as passive energy storage (67). However, recent studies have revealed they may have an important role in the tumour microenvironment and in breast cancer progression (15-17, 68, 69). An experimental co-culture system has been developed to grow breast tumour cells with adipocytes (15). This co-culture alters both adipocyte and breast tumour cell phenotypes. Adipocytes become dedifferentiated, less lipid rich, and secrete factors which enhance survival and migration of breast tumour cells (15, 17, 68). These cells are called cancer-associated adipocytes (CAA). Breast cancer cells co-cultured with CAA become more resistant to radiation therapy (16) and chemotherapy (our labs own unpublished data), show enhanced growth (70), and are more migratory and invasive (Figure 1.1) (15, 17, 68, 71). Various adipocyte secreted factors have been implicated in promoting this aggressive phenotype in breast cancer cells. For example, CAA-secreted proteases, MMP-11 and PAI-1 (plasminogen activator inhibitor-1) (68, 69, 72), and inflammatory cytokine IL-6 have been suggested as key factors in the acquired resistance to radiation therapy observed in breast cancer cells (15). However, little is known about tumour cell metabolism in breast cancer cells that have been exposed to CAAs. 13 Figure 1.1: Interactions between tumour cells and mature adipocytes as described in text. Adapted from Wang et al. (73). Reprinted with permission from Elsevier. 1.4 Metabolism in cancer Metabolism has been detailed as an emerging hallmark of cancer and is thought to be important for cancer development (74, 75). Tumour growth and survival are also supported by changes in fundamental oncogenic pathways which alter tumour metabolism (76). Metabolism has an important role in cancer cells as it is the basis of ATP generation, biosynthesis, and maintenance of redox status (76). Alterations in these pathways can create the support necessary for the rapid proliferation and growth of a tumour. Early metabolic research in cancer was pioneered by Warburg and primarily focused on glycolysis. Since then research has expanded and many pathways in metabolism have been revealed to have important oncogenic roles (75, 76). The tumour microenvironment has been implicated as having an important role in these metabolic changes (14, 77). 1.4.1 Warburg theory Otto Warburg, a German biochemist, in the 1920s proposed that even when oxygen is present cancer cells use glycolysis for ATP synthesis (78, 79). This was called aerobic glycolysis and the phenomenon became known as the “Warburg Effect” (80). Warburg posited that this was due to the oxidative phosphorylation (OXPHOS) pathway being disrupted by mitochondrial dysfunction, which led cancer cells to switch from OXPHOS 14 to glycolysis to produce sufficient ATP (80, 81). His theory has been challenged in more recent years as it has been found that in many cancer cells, although there is upregulated aerobic glycolysis, OXPHOS is not disrupted (76, 80, 81). This upregulated glycolysis may be helping to compensate for high biosynthetic demands to keep up with the proliferation rate of cancer cells whilst reducing oxidative stress (80). It is likely that these changes seen in cancer cells are due to mutations in oncogenes and tumour suppressors that have an effect on metabolism (76). 1.4.2 Two-compartment tumour metabolism Not all cancer cells show the metabolic phenotype associated with the Warburg effect (17, 82). Pavlides and colleagues (2009) discovered that the stromal fibroblasts in breast tumours can be induced, via oxidative stress from the cancer cells, to undergo aerobic glycolysis (63). This results in the secretion of metabolites creating an energy-rich microenvironment for tumour growth; they termed this the “Reverse Warburg Effect” (63). This holds true to Warburg’s original observations that tumours undergo aerobic glycolysis, the glycolysis is just occurring in the tumour microenvironment compartment; specifically the CAFs (18). Through further in vitro studies they expanded this with a parallel model called the autophagic tumour stroma model of cancer, in which cancer cells are fuelled by stromal cells undergoing autophagy (18). Autophagy is a pathway involved in self-degradation of organelles and proteins and therefore can have a role in cancer through provision of nutrients and prevention of cell death (83, 84). This metabolic coupling is termed “twocompartment tumour metabolism” where the catabolic stromal compartment undergoes aerobic glycolysis and autophagy which drives the anabolic growth of tumours in the cancer cell compartment via oxidative metabolism (19). This study was conducted in CAFs and breast cancer cells. The glycolytic CAFs produce L-lactate and ketone bodies, which are high energy nutrients that can be utilised by the neighbouring breast cancer cells for ATP production (46). They also produce reactive oxygen species (ROS) that promote mutations in the cancer cells to drive tumour-stroma co-evolution (46). Previous studies, and our own, have reported reduced size and loss of lipids in CAAs (15, 17). It is possible that this two-compartment tumour metabolism model is analogous to interactions between CAAs and breast cancer cells. This would be observed as adipocytes 15 undergoing lipolysis and the produced glycerol and fatty acids being metabolised by the breast cancer cells. 1.4.3 β-oxidation β-oxidation is often an overlooked metabolic pathway in studies of cancer metabolism. Fatty acids are a good source of fuel for ATP production and they also provide substrates to counteract oxidative stress (85). Fatty acids may be present in the tumour microenvironment and therefore β-oxidation should be considered as a potential pathway supporting the rapid proliferation and growth of cancer (17). β-oxidation is a catabolic bioenergetic process that occurs in peroxisomes and mitochondria (85). It involves the breakdown of long chain fatty acids into acetyl-CoA, NADH and FADH2 in a cyclical 4-step sequence (85, 86). The NADH and FADH2 subsequently enter the electron transport chain where they produce ATP (Figure 1.2) (85). Acetyl-CoA enters the Kreb’s Cycle and one of the products is citrate; citrate in the cytoplasm can generate NADPH (85). NADPH is capable of counteracting oxidative stress (85); this is important in cancer cells to balance the ROS for cancer cell survival (87). ROS are highly reactive radicals that are capable of damaging various cellular components by oxidation reactions (88). This promotes genetic instability so cancer cells can mutate and ‘evolve’ survival mechanisms more readily (87). ROS are produced by cancer cells under oxidative stress. Almost all types of cancer are found to have increased ROS levels which promote tumour development and progression (88). 16 Figure 1.2: Key proteins and reactions in the fatty acid oxidation pathway as described in the text. Adapted from (85). ACC, acetyl-CoA carboxylase; AMPK, AMP kinase; CPT1, carnitine palmitoyl transferase 1; TCA, the citric acid cycle. Fatty acid synthesis is a well-studied topic in cancer research but fatty acid catabolism is largely ignored. Fatty acids need to be synthesised in proliferating cells to build cell membranes and various cellular components such as lipid signalling molecules (89). Tumour cells synthesise a large proportion of their fatty acids de novo regardless of available dietary lipids (89). Fatty acid synthase (FASN), the key enzyme in fatty acid synthesis, is expressed at increased levels in cancer cells, including breast cancer, and is associated with poorer prognosis (90, 91). Fatty acids can therefore be endogenous or exogenous but both can be used in β-oxidation or synthesis (85, 89, 91-93). This shows that fatty acids are already an important factor in cancer growth and may have a role not only in synthesis but also metabolism. As outlined above research has mostly focused on glycolysis due to the “Warburg Effect” and its prevalence in tumours. However recent studies have begun to investigate βoxidation in cancer cells. Fatty acid oxidation was observed by Liu in prostate cancer (82). Low levels of glycolysis are common in prostate cancer and upon further analysis it was found that prostate cancer cells have increased β-oxidation to provide the ATP for rapid growth and proliferation (82). Although two-compartment tumour metabolism 17 originally looked at fibroblasts and breast cancer cells it has been extended by Nieman and colleagues (2011) to adipocytes and ovarian cancer cells (17). They showed that lipids were transferred to ovarian cancer cells from omental adipocytes via induced adipocyte lipolysis. This was followed by β-oxidation of the transferred fatty acids in the mitochondria of the cancer cells (Figure 1.3) (17). They concluded that the fatty acids provided by adipocytes were used to fuel rapid tumour growth and the associated migration and invasion that results in metastasis. Figure 1.3: Summary of metabolic changes that occur in interacting ovarian cancer cells and adipocytes as described in the text (94). Reprinted with permission from Nature Publishing Group. This expands our understanding of how adipocytes in the tumour microenvironment may affect tumour growth and spread. It is possible that β-oxidation is altered in breast cancer to take advantage of fatty acids released by lipolysis from surrounding adipocytes. Breast cancer cells may also be signalling the breast adipocytes to release the fatty acids by lipolysis so they can be used as fuel or building blocks for tumour growth, invasion and metastasis. 1.4.4 Proteins involved in β-oxidation 1.4.4.1 Acetyl-CoA carboxylase (ACC) Fatty acid oxidation has various rate-limiting enzymes one of which is acetyl-CoA carboxylase (ACC); a key regulator of fatty acid oxidation (95, 96). ACC catalyses the reaction that converts acetyl-CoA to malonyl-CoA which determines levels of fatty acid synthesis and metabolism (Figure 1.2) (95). Malonyl-CoA is an inhibitor of carnitine 18 palmitoyl transferase 1 (CPT1); an enzyme that moves fatty acids across the mitochondrial membrane for β-oxidation (86, 97). As a regulatory enzyme in fatty acid metabolism, ACC is thought to have a role in obesity and may also have a role in cancer (98). AMP-activated protein kinase (AMPK) functions in regulating energy balance in cells (99). This is achieved by sensing cellular ATP levels and altering the rate of ATPsynthesis and ATP-consuming pathways to balance cellular needs (99, 100). This is important in cell survival and as such AMPK is a critical enzyme that may be altered in cancer (101). Stress signals from the tumour microenvironment can activate AMPK and play a role in activating lipid metabolism pathways (101). When AMPK is activated it phosphorylates and inhibits ACC which reduces malonyl-CoA levels and therefore increases CPT1 activity; increasing β-oxidation (Figure 1.2) (96, 102). Due to its effects on ACC, AMPK is a key regulator of fatty acid metabolism and synthesis (101). Both AMPK and ACC are thought to be linked to obesity, and phosphorylation by leptin and adiponectin stimulates fatty acid oxidation (96, 103). 1.4.4.2 Carnitine palmitoyl transferase 1 (CPT1) CPT1 is an important transmembrane protein in fatty acid oxidation (104, 105). It is located in the outer mitochondrial membrane (106) and creates acylcarnitine by catalysing the transfer of acyl-CoA from a long-chain acyl-CoA ester to carnitine (86). Acylcarnitine can be shuttled across the mitochondrial membrane making it available for β-oxidation in the mitochondrial matrix (86). There is potential for CPT1 to have an oncogenic role in cells due to its rate limiting role (104). When CPT1 activity is increased this allows increased movement of fatty acid products into the mitochondria for β-oxidation (85). CPT1 has already been associated with an oncogenic effect in breast cancer cells (104). Linher-Melville and colleagues (2011) showed that breast cancer cells, but not normal breast epithelial cells, stimulated with prolactin had increased CPT1 expression and activity (104). They suggested that the cancer cells may use increased CPT1 activity to provide fuel for their high energy demands. If the surrounding adipocytes are indeed undergoing lipolysis and releasing free fatty acids, an increase in CPT1 activity could allow the utilisation of this alternate energy source as a fuel for the cancer cells. 19 1.4.4.3 Mitochondrial glycerol-3-phosphate dehydrogenase (mGPDH) mGPDH, also known as GPD2, is the simplest component in the glycerophosphate shuttle (107). It has two alternate isoforms produced by alternative splicing; 81kDa and 68kDa (108). The glycerophosphate shuttle connects the cytoplasmic and mitochondrial metabolic pathways and allows glycolysis produced NADH to contribute to oxidative phosphorylation (109). Thus, mGPDH functions at the intersection of oxidative phosphorylation, glycolysis, and fatty acid oxidation (110). Altering the function of mGPDH has been shown to affect cell metabolism and energy levels (110). mGPDH can be regulated by free fatty acids and may be involved in fatty acid oxidation pathways (110). mGPDH produces ROS which explains why it is suppressed in most tissues but detected at an increased level in cancer cells (111-113). Prostate cancer cell lines are found to have elevated mGPDH levels and activity compared to normal prostate epithelial cell lines (111). Due to the increased glycolysis also observed in these prostate cancer cells it is hypothesised that mGPDH plays a role in sustaining the high rate of metabolism required for prostate cancer growth and proliferation (111). It has also been observed that some breast carcinomas have increased mGPDH activity (114), and thus, it is a protein of interest in metabolic pathways. 1.5 Summary and hypothesis Previous research in the Mackenzie Cancer Research Group (MCRG) has shown that adipocyte breast cancer co-culture leads to increased resistance to chemotherapy agents (Doxorubicin and Paclitaxel) in both oestrogen receptor positive (MCF7) and hormone receptor negative (MDA-MB-231) breast cancer cell lines. Furthermore, both MCF7 and MDA-MB-231 breast cancer cells become more highly motile after co-culture with human breast adipocytes. However, as yet, nothing is known about breast cancer cell metabolism after co-culture with human breast adipocytes. Previous studies in breast cancer have already noted the reverse Warburg effect and autophagy in the tumour stroma leading to the two-compartment tumour metabolism model (18, 46, 63). These models have mostly overlooked the presence of adipocytes in the breast tumour stroma. Only a few studies have investigated adipocyte effects on cancer cell metabolism, and breast cancer was not one of the examined cancers (17, 82). No previous study has looked at how the interactions between human breast adipocytes and breast cancer cells may be affecting cancer cell metabolism. Adipocytes may be 20 providing the fuel for glycolysis and β-oxidation in the breast cancer cells allowing growth, invasion, and metastasis. In this project, the aim is to investigate whether adipocytes alter the metabolism of tumour cells when they are cultured together. This will give greater insight into the biology of adipocytes and the interactions between breast tumour cells and this largely neglected stromal cell type. It is hypothesised that breast tumour cells are capable of activating lipid release by nearby adipocytes, and may metabolise the resulting glycerol and fatty acids to fuel migration and invasion (Figure 1.4). Breast cancer cell Cancer-Associated Adipocyte Induce Lipolysis β-oxidation lipolysis Migration Glycerol, Invasion Fatty acids Figure 1.4: Hypothesis of breast cancer-adipocyte interactions. The induction of lipolysis in cancerassociated adipocytes by breast cancer cells results in release of nutrients. Glycerol and fatty acids can then be up taken by the breast cancer cells and used to fuel migration and invasion. 1.6 Aims and objectives To observe metabolism in breast tumour cells and look at potential changes when the breast tumour cells are co-cultured with adipocytes. Specific objectives are outlined below: 1. Tumour cell metabolism will be characterised in breast cancer cell lines MCF7 (ER+) and MDA-MB-231 (ER-/PR-/HER2-). Furthermore tumour cell metabolism will be investigated in MCF7 and MDA-MB-231 cells grown in adipocyte co-culture. This will investigate if breast cancer cells are using glycerol and fatty acids for tumour cell glycolysis, β-oxidation and mitochondrial metabolism. 2. β-oxidation of fatty acids may be an important source of energy for tumour cell migration and invasion. This will be analysed by determining the levels of tumour cell cytosolic and mitochondrial proteins that are involved in metabolism. 21 3. Adipocytes and breast tumour cells will be co-cultured, and adipocyte-derived glycerol and free fatty acids will be measured in the resulting cell culture media. This will also allow analysis of whether adipocytes are induced to undergo lipolysis by breast tumour cells. Due to time restrictions in this project, these objectives were reduced. Alterations are listed below: 1. Investigation of tumour cell metabolism in cells grown in adipocyte co-culture will only be undertaken in MCF7 cells. 2. Only adipocyte-derived glycerol will be measured in the adipocyte-breast cancer cell co-culture media. 22 2 Materials and Methods 2.1 Breast adipose tissue samples and processing 2.1.1 Collection of patient samples Breast adipose tissue samples were collected from female patients at Christchurch Hospital undergoing surgery for therapeutic mastectomy, prophylactic mastectomy and breast reductions. Ethical approvals were obtained for banking adipose tissue via the Cancer Society Tissue Bank Christchurch (Upper South Ethics Committees, 05/11/09), and collection of human adipose tissue for short-term culture experiments (UO Human Ethics Approval 12/319). All patients gave their informed written consent. Patients were from 25-85 years of age. A total of 15 adipose tissue samples were collected over the duration of this study (April – August, 2015). 2.1.2 Isolation of pre-adipocytes Protocol adapted from Lee et al. (2005) (115). Adipose tissue samples were processed by removal of fibrous and vascular areas. The adipose tissue was minced, before digestion by collagenase (1 mg/ml in HBSS). Collagenase was used at a ratio of 1ml collagenase to 1 g adipose tissue. Collagenase mix was incubated at 37°C for 2 hours with shaking every 15 minutes. An equal amount of pre-adipocyte media (Appendix 5.1.1) was added to collagenase-adipose mix. Sample had fibrous tissue removed using a metal sieve and was then transferred to a 50 ml tube (BD FalconTM) and centrifuged for 10 minutes at 2500 x g (Multifuge® 1 S-R, Heraeus) at room temperature. Top fraction was discarded; this left the remaining pre-adipocyte fraction as a pellet. The pellet was resuspended in 10 ml pre-adipocyte media and sieved using a cell sieve (Cell strainer 70µm nylon, BD Falcon™) into a new 50 ml tube. Sample was spun again for 10 minutes at 2500 x g at room temperature. Supernatant was discarded and cell pellet resuspended in 1 ml of preadipocyte media per 0.6 g adipose tissue. Suspension was plated 1 ml/well into 12-well CellBIND® plates (Costar®, Corning) and incubated at 37°C with 5% CO2 (Function Line, Heraeus). 2.1.2.1 Removal of red blood cells The day after plating, 1 ml of 1X Phosphate Buffered Saline (PBS) (Appendix 5.2.1) was added to each pre-adipocyte well. PBS was pipetted across the bottom of the well to 23 remove red blood cells. Media was then removed and replaced with 1X PBS and the wash was repeated. PBS was removed and replaced with 1 ml/well pre-adipocyte media. 2.1.3 Culture of pre-adipocytes Pre-adipocytes were grown in pre-adipocyte media which was changed every 2-3 days. They were observed using light microscopy (CK40, Olympus) until they reached 100% confluence. Differentiation treatment began two days after confluence was observed. Pre-adipocyte media was replaced with 1ml/well serum free media (Appendix 5.1.2) 24 hours prior to co-culture. 2.1.4 Differentiation of pre-adipocytes into mature adipocytes Differentiation of pre-adipocytes to mature adipocytes took 14-21 days. Serum free media was removed and collected (section 2.6.1). Pre-adipocytes were differentiated for 5 days in 1 ml/well differentiation media (Appendix 5.1.3), which was then replaced with 1 ml/well differentiation media minus IBMX (Appendix 5.1.3) for a further 5 days. This media was subsequently replaced with 1 ml/well maintenance media (Appendix 5.1.4) which was renewed every 3-4 days. Cells were monitored for maturation by light microscopy. Maturation was observed as lipid droplets accumulating and filling the adipocytes (Supplementary Figure 5.1). Typically, cells reached maturation at approximately 7 days after maintenance media was added, and were considered ready for experiments when mature adipocyte confluence was >30% of the plate area. 2.2 Breast cancer cells The human breast cancer cell lines MCF7 and MDA-MB-231 were obtained from American Tissue Culture Collection (Cryosite Distribution Pty. Ltd., Australia). The MCF7 cell line (ATCC® HTB-22™) is an oestrogen receptor positive (ER+) human breast epithelial cell adenocarcinoma that was acquired from a 69 year old Caucasian female. The MCF7 cell lines is classified as molecular subtype luminal A and is more receptive to hormonal treatment due to its ER+ status (116). The MDA-MB-231 cell line (ATCC® CRM-HTB-26™) is commonly referred to as a triple negative breast cancer cell line. It is an oestrogen, progesterone, and human epidermal growth factor receptor negative (ER-/PR-/HER2-) human breast epithelial cell adenocarcinoma acquired from a 51 year old Caucasian female, and is classified as claudin-low (116). Claudin-low cell lines have mesenchymal and stem cell-like features that cause attraction of stromal cells into the microenvironment (35). Hormonal treatment is less effective in MDA-MB-231 24 due to its triple negative hormone receptor status (116). These cell lines were chosen because of these different hormone receptor profiles. In summary, MDA-MB-231 is a very invasive and aggressive cell line in comparison to MCF7 (117-119). Cell lines were grown in BD FalconTM tissue culture flasks (75 cm2 and 25 cm2) in breast cancer cell media (Appendix 5.1.5) and incubated at 37°C with 5% CO2. Cells were stored in liquid nitrogen until used. 2.2.1 Passage Cells were passaged every 2-3 days up to passage 30 to minimise the effects of phenotypic drift on the cell lines. Media was removed from the flask and cells washed with 1X PBS before addition of TrypLE (TrypLE™ Express, Gibco®, Thermo Fisher Scientific). Cells were incubated with TrypLE at 37°C for 3-5 minutes and gently agitated at which stage they were observed to be detached from the flask. Cells were then transferred to a 15 ml tube and spun for 3 minutes at 500 x g (Centrifuge 5702, Eppendorf). Supernatant was discarded and cells resuspended in breast cancer cell media. Appropriate quantities of cells were added to a new 25 cm2/75 cm2 tissue culture flask with 6 ml/12 ml of media respectively. MCF7s were split 1/6 and MDA-MB-231s 1/8. 2.3 Co-culture of mature adipocytes and breast cancer cell lines 2.3.1 Day 1 Protocol adapted from Dirat et al. (2011) (15). Maintenance media on mature adipocytes was replaced with 1ml/well serum free media (Appendix 5.1.2) 24 hours prior to coculture. Transwell inserts (Transwell® permeable supports 0.4 µm polyester membrane 12 mm insert, Costar®, Corning) were soaked in serum free media at 37°C for ≥1 hour. Mature adipocytes to be used were photographed for the record; examples can be seen in Appendix 5.3.1. Serum free media on mature adipocytes was removed and collected (section 2.6.1) and replaced with 1 ml/well serum free media. Breast cancer cells were trypsinised and counted using a haemocytometer before being diluted to desired concentrations in serum free media: MCF7 2.4x105 cells/ml, MDA-MB-231 2x105 cells/ml. Serum free media was removed from Transwell inserts and inserts were gently placed into mature adipocyte wells. Breast cancer cells were added in 500 µl amounts to inserts at a final concentration of MCF7 1.2x105 cells/insert, MDA-MB-231 1x105 cells/insert. Co-culture plates were then incubated at 37°C with 5% CO2 for 3 days. 25 2.3.2 Day 4 Serum free media was removed from Transwell inserts and mature adipocyte wells and collected (section 2.6.1). Transwell inserts were moved to a new 12-well plate (Falcon) and 200 µl TrypLE was added to each insert. After 3-5 minute incubation at 37°C breast cancer cells were collected and spun at 500 x g for 3 minutes. Supernatant was discarded, cells were resuspended in 1X PBS and the centrifuge was repeated. Cells were then treated for use in either Seahorse experiments (section 2.4) or Western blotting (section 2.5). Mature adipocytes were washed twice with 1X PBS before addition of 1ml/well serum free media. Mature adipocytes were photographed to record any visible changes. Mature adipocytes were incubated at 37°C with 5% CO2 for 24 hours. 2.3.3 Day 5 Serum free media was removed from mature adipocyte wells and collected (section 2.6.1). 2.4 Seahorse XF24 Extracellular Flux Analyser XF24 Extracellular Flux Analyser (Seahorse Biosciences) was used to assay mitochondrial function and fatty acid utilisation of intact cells. The XF analyser uses solid state probes embedded on a sensor plate to detect the concentration of oxygen and protons in the media of cells cultured in a complementary microplate. Repeat measurements (typically 2 – 5 minutes) are made in which the sensor cartridge is lowered to create a sealed microchamber (7 µl) above the cell monolayer. The XF analyser then measures oxygen consumption rate (OCR) as picomoles per minute (pmol/min) and extracellular acidification rate (ECAR) as milli pH per minute (mpH/min). These measures give indications of oxidative respiration and glycolysis, respectively. After each measurement, the sensor cartridge is lifted allowing cells to re-equilibrate in the excess volume of media before the next measurement cycle. The non-destructive measurements made by the XF system allow for measuring over extended periods of time and with the ability to use cells for post-XF analysis assays such as total protein determination. Additionally an integrated drug port delivery system allows for the sequential injection of up to four compounds to cell wells during assays to create metabolic profiles (120, 121). 26 2.4.1 Optimisation for XF24 To begin work with the new cell lines MCF7 and MDA-MB-231 on the Seahorse XF24 Extracellular Flux Analyser, optimisation procedures must be undertaken. Cell seeding density was optimised to produce an even monolayer of cells at the bottom of an XF 24well plate. MCF7 and MDA-MB-231 cells were grown to 80% confluence before being seeded in triplicate at densities ranging from 1x104 – 5x105 cells/well in an XF 24-well plate; the seeding density range was determined from previous literature (122-129). Cells were grown overnight (20 hours) at 37°C with 5% CO2 and their confluence was then evaluated using light microscopy (section 3.1.1). This followed the established protocol for non-adherent cells published by Seahorse Biosciences (130). The concentrations of drugs to be injected during the mitochondrial stress test (section 2.4.2) were optimised in each cell type for maximal effect (section 3.1.1). 2.4.1.2 MTT assay A parallel experiment was run in a standard 96-well plate (same growth surface area as an XF 24-well plate) to confirm uncompromised viability at various densities. Cells in the 96-well plate (section 2.4.1) underwent an MTT assay to measure cell viability at various seeding densities to ensure cells were growing optimally Media was removed from all wells and replaced with 100 µl MTT assay media (Appendix 5.1.6) and the plate was incubated at 37°C for 4 hours. Following this, 100 µl of solubilisation solution (Appendix 5.2.3) was added to each well and pipetted up and down. Absorbance was measured at 544 nm in Victor3 (1420 multilabel counter, Perkin Elmer™). 2.4.2 Mitochondrial stress test The mitochondrial stress test is designed to interrogate the respiratory capacity of mitochondria enabling the investigator to create a bioenergetic profile of the cells of interest. Electron transport chain activity is disrupted through sequential addition of oligomycin (ATP synthase inhibitor), FCCP (carbonyl cyanide-p- trifluoromethoxyphenylhydrazone) (uncoupler), and antimycin-A (complex III inhibitor) with corresponding changes in OCR thereby, enabling the calculation of bioenergetic parameters that can be compared between cell lines or control and treatment groups (section 3.1) (121). Sensor cartridges were hydrated with 1 ml/well calibrant solution (Seahorse Biosciences) in a 37°C non-CO2 incubator overnight. MCF7 and MDA-MB-231 cells were seeded at 27 4x104 cells/well for both cell lines in 100 µl breast cancer cell media in an XF 24-well plate. Co-cultured MCF7 cells were seeded at 4x104 – 6x104 cells/well in 100 µl serum free media in an XF 24-well plate. This increase in seeding density was to account for differences produced by co-culture which resulted in slower adherence and growth. Seeded plates were left at room temperature for 30-60 minutes before incubation at 37°C with 5% CO2 to aid even adherence. After 4 hours of incubation at 37°C, 150 µl/well of additional media was applied before overnight incubation at 37°C with 5% CO2. After 24 hours the seeded cells underwent media transfer and 600 µl of mitochondrial stress test assay media (Appendix 5.1.7) remained per well. The plate was then incubated at 37°C in a non-CO2 incubator for 30 minutes. Compounds oligomycin (Sigma-Aldrich), FCCP (Sigma-Aldrich), and antimycin-A (Sigma-Aldrich) were prepared for injection. Oligomycin was prepared to a final concentration of 1 µM after injection. FCCP was prepared to a final concentration of 0.5 µM for MCF7 or 1 µM for MDA-MB-231 after injection. Antimycin-A was prepared to a final concentration of 10 µM after injection (section 2.4.1 and 3.1.1 for drug optimisation). Injection ports were loaded with the appropriate compounds and unused ports were loaded with mitochondrial stress test assay media. The plate was loaded into the Seahorse XF machine and the following protocol was run (Table 2.1): Table 2.1: Mitochondrial stress test protocol Command Calibrate Mix, Wait, Measure Inject Oligomycin Mix, Wait, Measure Inject FCCP Mix, Wait, Measure Inject Antimycin-A Mix, Wait, Measure Number of loops Times (minutes) 7 3, 2, 3 3 3, 2, 3 3 3, 2, 3 3 3, 2, 3 2.4.3 Normalisation of XF data Total protein content of cells in each well of a XF 24-well plate was quantified using a BCA protein assay. A standard curve of bovine serum albumin (BSA) was prepared using a dilution series in 1X PBS creating a working range of 0-2000 µg/ml. BSA curve and controls were pipetted in 25 µl aliquots into a 96-well plate in duplicate. XF 24-well plate had media removed and 25 µl lysis buffer (Appendix 5.2.4) was added to each well. Working reagent was prepared at 50:1 (reagent A: reagent B); 200 µl was added to each well on the 96-well plate and the XF 24-well plate. Plates were mixed on shaker for 30 28 seconds, protected from light, and incubated at 37°C for 30 minutes. XF 24-well plate samples were transferred to the 96-well plate. Absorbance was measured at 544 nm using Victor3 plate reader. The total protein content of each well was calculated and then entered into the normalisation function of the Seahorse Wave Desktop 2.2 software. 2.4.4 Fatty acid oxidation test optimisation The fatty acid oxidation test is designed to determine the ability of cells to oxidise endogenous and exogenous fatty acids. It follows the same injection profile as the mitochondrial stress test during the assay. Prior to the assay wells are treated with palmitate and etomoxir (Sigma-Aldrich). This allows cells to be driven to oxidise exogenous and endogenous fatty acids and the contribution to respiration can be calculated (section 3.1.6) (131). Sensor cartridges were hydrated with 1 ml/well calibrant solution in a 37°C non-CO2 incubator overnight. Cells were seeded at 4x104 cells/well for both MCF7 and MDAMB-231 cells in 100 µl substrate limited media (Appendix 5.1.8) in an XF 24-well plate. After 4 hours of incubation at 37°C, 150 µl/well of substrate limited media was applied before overnight incubation. After 24 hours the seeded cells underwent media transfer and 495 µl of FAO assay media (1X) (Appendix 5.1.9) remained as final volume per well. The plate was then incubated at 37°C in a non-CO2 incubator for 30 minutes. Compounds oligomycin, FCCP, and antimycin-A were prepared for injection. Oligomycin was prepared to a final concentration of 1 µM after injection. FCCP was prepared to varying concentrations (section 2.4.3.1) after injection. Antimycin-A was prepared to a final concentration of 10 µM after injection. Injection ports were loaded with the appropriate compounds and unused ports were loaded with FAO assay media (1X). Directly before loading, 105 µl of BSA or BSA:palmitate was applied to selected wells. The plate was loaded into the Seahorse XF machine and the following protocol was run (Table 2.2): 29 Table 2.2: Fatty acid oxidation stress test protocol Command Calibrate Mix, Wait, Measure Inject Oligomycin Mix, Wait, Measure Inject FCCP Mix, Wait, Measure Inject Antimycin-A Mix, Wait, Measure Number of loops Times (minutes) 5 3, 2, 3 3 3, 2, 3 3 3, 2, 3 3 3, 2, 3 2.4.3.1 Optimisation of FCCP for fatty acid oxidation stress test The fatty acid oxidation stress test was carried out with final experiment levels of BSA or BSA:palmitate but with no etomoxir for MCF7 cells over a range of FCCP concentrations (0.5 µM, 1 µM, 1.5 µM, 2 µM, 3 µM, 4 µM, 5 µM). The results were then analysed to select a FCCP concentration (section 3.1.6.2). 2.4.3.2 BSA:palmitate conjugation BSA:palmitate conjugation must be undertaken so that palmitate, the fatty acid, can be taken up by the cells. Conjugation was completed according to Seahorse Bioscience protocol (132). Firstly 226.7 mg fatty acid poor BSA (Gibco®, Thermo Fisher Scientific) was added to 10 ml NaCl (150 mM) in a pre-warmed beaker with stir bar. This beaker was covered with parafilm and placed in a water bath heated to 37°C and stirred till BSA was completely dissolved. The BSA was then filtered and 5 ml was diluted with an additional 5 ml NaCl (150 mM) to create a 0.17 mM stock. This stock was aliquoted into glass vials and stored at -20°C. The remaining 5 ml was kept at 37°C for the next part of the conjugation. Next, 30.6 mg of sodium palmitate (Sigma-Aldrich) was added to 44 ml NaCl (150 mM) in an Erlenmeyer flask. This was covered with parafilm and placed in a water bath and gradually heated to 70°C. For conjugation 4 ml of palmitate solution was transferred to the BSA solution at 37°C. This was held at 37°C and stirred constantly for 1 hour. Final solution was adjusted to 10 ml with NaCl (150mM) and adjusted to pH 7.4. This created a 0.17 mM BSA:1mM palmitate conjugation. This was aliquoted into glass vials and stored at -20°C. 30 2.5 Western Blot 2.5.1 Cell lysis 2.5.1.1 MCF7 and MDA-MB-231 cells cultured alone Breast cancer cells had media removed and were washed twice with 1X PBS. Cells were moved to ice and lysed with lysis buffer. Cells were scraped off the plastic using a modified Pasteur pipette. Lysate was transferred to pre-chilled tubes and rotated at 4°C for 10 minutes. Samples were then sonicated on ice with the probe sonicator (OmniRuptor 4000, OMNI International Inc.) using 3 pulses at power 30. Samples were spun at 18000 x g (Centrifuge 5417R, Eppendorf) for 10 minutes at 4°C. The supernatant was then transferred to a pre-chilled tube and stored immediately at -20°C. 2.5.1.2 MCF7 and MDA-MB-231 cells obtained from adipocyte co-culture Breast cancer cell pellets obtained from co-culture (section 2.3.2) were resuspended in lysis buffer and rotated at 4°C for 10 minutes. Samples were then sonicated on ice with the probe sonicator using 3 pulses at power 30. Samples were spun at 18000 x g for 10 minutes at 4°C. The supernatant was then transferred to a pre-chilled tube and stored immediately at -20°C. Final co-culture samples for MCF7 and MDA-MB-231 were cultured with three biologically separate adipocyte samples. 2.5.2 Protein quantification Total protein content of samples was quantified using a BCA protein assay. A standard curve of BSA was prepared using a dilution series in 1X PBS creating a working range of 0-2000 µg/ml. BSA curve, controls and unknown samples were pipetted in 25 µl aliquots into a 96-well plate in triplicate. Working reagent was prepared at 50:1 (reagent A: reagent B) and 200 µl was added to each well. Plate was mixed on shaker for 30 seconds, protected from light, and incubated at 37°C for 30 minutes. Absorbance was measured at 544 nm in Victor3. 2.5.3 SDS-PAGE Samples were prepared so that 5-20 µg of protein was loaded per lane. From each sample 50 µl (2 mg/ml) was added to 50 µl reducing sample loading buffer (Appendix 5.2.6). Samples were heated to 70°C for 10 minutes, cooled to room temperature, and mixed by vortex. Gels were NuPage 4-12% 12-well (Novex). Control ladders used were SeeBlue® Plus2 standard (Novex) and Precision Plus Protein™ standard (Bio-Rad). Precision Plus 31 Protein™ standard was used for gels that were undergoing Western blotting for large proteins (265-280kDa) as the ladder has a larger band to compare to. Samples were loaded into the gel and run in running buffer (Appendix 5.2.7). Gels that were blotted for large proteins were run at 180V for 100 minutes. Gels that were blotted for small-medium proteins were run at 160V for 70 minutes. 2.5.4 Transfer Proteins were transferred onto PVDF membranes. The membrane was pre-soaked in methanol for 1 minute, rinsed in H2O and soaked in transfer buffer (Appendix 5.2.8). The Western blot was run at 25V for 60 minutes. The membrane was blocked for 1 hour with 5% w/v low fat milk in 1X TBS-T (Appendix 5.2.9) (for anti-CPT1A, anti-GPD2, antiβ-actin) or in 5% w/v BSA in 1X TBS-T (for anti-ACC1 and anti-phospho-ACC). Membranes were then cut at 50kDa; top sections were probed for CPT1A, ACC1, phospho-ACC or GPD2, bottom sections were probed for β-actin. 2.5.5 Antibodies Antibodies were optimised with preliminary experiments to obtain the final concentrations. Primary antibodies (Table 2.3) were applied to the membrane in 10 ml quantities and incubated overnight at 4°C with gentle shaking. Primary antibodies were kept at -20°C and reused up to 5 times. Membranes were equilibrated to room temperature and washed in 1X TBS-T for 10 minutes; this was repeated 3 times. Secondary antibodies (Table 2.3) were diluted in 1X TBS-T and applied to the membrane in 10 ml quantities. Membranes were incubated with secondary antibodies at room temperature for 1 hour. 32 Table 2.3: Primary and secondary antibody concentrations used in Western blotting Primary antibody Antibody Concentration (catalogue number) Diluent (in 1X TBS-T) Manufacturer Secondary Antibody (catalogue number) Manufacturer CPT1A mouse monoclonal antihuman/mouse/rat 1:5000 (ab128568) 5% w/v low fat milk Abcam goat polyclonal anti-mouseHRP 1:15000 (P044701) Dako GPD2 rabbit monoclonal anti-human 1:1000 (ab182144) 5% w/v low fat milk Abcam goat polyclonal anti-rabbit-HRP 1:5000 (7074) Cell Signaling Technologies ACC1 rabbit monoclonal antihuman/mouse/rat 1:1000 (4190) 5% w/v BSA Cell Signaling Technologies goat polyclonal anti-rabbit-HRP 1:5000 (7074) Cell Signaling Technologies PhosphoACC rabbit monoclonal antihuman/mouse/rat 1:1000 (11818) 5% w/v BSA Cell Signaling Technologies goat polyclonal anti-rabbit-HRP 1:5000 (7074) Cell Signaling Technologies β-actin mouse monoclonal anti-β-actin 1:20000 (A5441) 5% w/v low fat milk Sigma-Aldrich goat polyclonal anti-mouseHRP 1:15000 (P044701) Dako 2.5.6 Detection Membranes were washed in 1X TBS-T for 10 minutes; this was repeated 3 times. Detection kit was AmershamTM ECLTM Prime Western Blotting Detection Reagent (GE Healthcare Life Sciences) following manufacturer instructions. This was followed by visualisation in the Alliance 4.7 (UVItec, Cambridge). 2.5.7 Re-probing After detection, membranes were washed in 1X TBS-T for 10 minutes. They were then stripped with mild stripping buffer (Appendix 5.2.10) for 5 minutes; this was repeated 2 times. Membranes were washed in 1X TBS-T for 5 minutes; this was repeated 2 times. Membranes were then washed in 1X TBS-T for 10 minutes; this was repeated 2 times. Membranes were then blocked (section 2.5.4), antibodies applied (section 2.5.5), and detection visualised (section 2.5.6). Membranes were stripped a maximum of 3 times. Proteins of interest were always probed for first, followed by β-actin. 33 2.6 Glycerol assay 2.6.1 Sample collection Media was collected from co-culture experiments at four time points: before differentiation, before co-culture, after 3 day co-culture, and 24 hours after co-culture. These conditioned media will henceforth be referred to as: pre-adipocyte, mature adipocyte, post co-culture, and 24 hours post co-culture, respectively. Post co-culture media was taken from both the Transwell inserts and the main well for combined analysis. All conditioned media were centrifuged at 500 x g for 5 minutes and 1 ml was collected and stored at -80°C. Pre-adipocyte conditioned media was collected from five different biological adipocyte cultures. Mature adipocyte conditioned media was collected from seven different biological adipocyte cultures. Post co-culture and 24 hours post co-culture conditioned media samples were collected from seven different biological adipocyte co-cultures with MCF7 cells and three different biological adipocyte co-cultures with MDA-MB-231 cells. 2.6.2 Detection Glycerol detection was performed using a free glycerol assay kit (Abcam). A standard curve for glycerol was prepared creating a working range of 0-200 µM. Samples, control, and standard curve were pipetted in 50 µl amounts into a 96-well plate in duplicate. Assay buffer, glycerol probe and glycerol enzyme mix were combined in the ratio 23:1:1 to create a reaction master mix. This was pipetted to create a 1:1 ratio of sample:reaction mix in standard and sample wells. The plate was then protected from light and incubated for 30 minutes at 37°C. Absorbance was measured at 544 nm in Victor3. 2.7 Statistical analysis All final analyses were performed using Microsoft Excel and the add-in, Data Analysis. 2.7.1 Seahorse XF analyser Data was gathered using Wave Controller 2.2 on the instrument. It was then analysed using Wave Desktop 2.2 and report generators; XF Mito Stress Test Report Generator V2 and XF Cell Energy Phenotype Test Summary Report Generator V2. Before analysis was conducted samples were normalised (section 2.4.3). MCF7 and MDA-MB-231 sample sets had their variance compared using F-tests. Sample sets of equal variance 34 were then tested for significance using an independent two-sample t-test assuming equal variances. Sample sets with unequal variance were tested for significance using an independent two-sample t-test assuming unequal variance. MCF7 cells grown alone were compared to adipocyte co-cultured MCF7 using a paired two-sample t-test for means. Significance was assumed when p<0.05. 2.7.2 Western blotting Western blot band density was calculated using ImageJ. Density was then normalised to β-actin before analysis. MCF7 cells and adipocyte co-cultured MCF7 cells had their sample set variance compared using F-tests. Sample sets of equal variance were then tested for significance using an independent two-sample t-test assuming equal variances. Sample sets with unequal variance were tested for significance using an independent twosample t-test assuming unequal variance. Significance was assumed when p<0.05. 2.7.3 Glycerol Assay Data was averaged across multiple experiments before analysis by one-way ANOVA. 35 3 Results 3.1 Analysis of metabolism in breast cancer cells and breast cancer cells cocultured with adipocytes using the Seahorse XF24 Extracellular Flux Analyser Cell lines were firstly optimised for use in the XF 24-well plates and for the mitochondrial stress test. The mitochondrial stress test was then carried out in MCF7 cells (n=2), MDA-MB-231 cells (n=4), and co-cultured MCF7 cells (n=2) and the results compared. Preliminary optimisation assays also began on the fatty acid oxidation stress test for MCF7 cells. 3.1.1 Optimisation for MCF7 and MDA-MB-231 cell lines on the Seahorse XF24 Extracellular Flux Analyser Cell seeding density was optimised in XF24 plates to ensure an even monolayer of cells with typical morphology (Figure 3.1: A), uncompromised viability, and OCR and ECAR rates within instrument recommendations. Cells seeded at 5x105/well and 1x105/well for MCF7 and MDA-MB-231 were too dense; cells were at >100% confluence and beginning to layer upon each other. However, MCF7 and MDA-MB-231 cells seeded at 2x104/well and 1x104/well were too sparse. They displayed abnormal morphology and some cells were not adhering (Supplementary Figure 5.2). MCF7 and MDA-MB-231 cells were also observed to be denser in the centre of the well than at the edge, for all densities (Supplementary Figure 5.3). To improve cell adherence and evenness of plating, Cell-Tak was trialled. Both MCF7 and MDA-MB-231 cells showed abnormal morphology when adhered using Cell-Tak. MCF7 cells had altered morphology with a mostly spherical appearance instead of the usual cobblestone appearance. MDA-MB-231 cells also did not have the expected elongated shape and were instead spherical (Figure 3.1: B). Cellular morphology was judged to be more important than adherence; Cell-Tak was not used for subsequent experiments. To achieve correct morphology and even plating, technique was modified. Cells were pipetted up and down in each well for even dispersal and after seeding, 24-well plates were left at room temperature for 30-60 minutes before incubation at 37°C. Cells were observed to be at ~85% confluency when plated at 4x104 cells/well for both cell lines and had more typical morphology (Figure 3.1: C). Results from an MTT assay concluded that neither MCF7 nor MDA-MB-231 cells were negatively affected when grown at this density (Supplementary Figure 5.4). Oxygen consumption rate of both cell lines under basal conditions seeded at 4x10 4 36 cells/well, was reviewed. MCF7 cells seeded at 4x104 cells/well produced an average basal OCR of 228 pmol/min and an average basal ECAR of 27 mpH/min. MDA-MB231 cells seeded at 4x104 cells/well produced an average basal OCR of 259 pmol/min and an average basal ECAR of 26 mpH/min. The XF output for both cell types fell within the basal ranges recommended by Seahorse Bioscience for OCR (50-400 pmol/min) and ECAR (20-120 mpH/min), respectively (133). 37 MDA-MB-231 Standard seeding 80% confluence MCF7 A) C) Final XF 24-well seeding 4 x 104 Cell-Tak seeding 5 x 104 B) Figure 3.1: Comparison of MCF7 and MDA-MB-231 cell seeding using representative 10X photographs. A) Standard culture flask seeding of MCF7 and MDA-MB-231 at 80% confluence. B) Cell-Tak seeding of MCF7 and MDA-MB-231 at 5x104 cells/well in an XF 24-well plate. C) Final seeding of MCF7 and MDAMB-231 in an XF 24-well plate at 4x104 cells/well. 38 Next, the response of each cell type to drugs used in the mitochondrial stress test was optimised. Oligomycin is a drug which prevents ATP synthesis by inhibiting ATP synthase (Figure 3.2) (121). This can be visualised on the XF analyser as a drop in OCR after the first injection (Figure 3.3: A, C). Previous literature indicated that oligomycin concentration required for maximal ATP synthase inhibition in MCF7 and MDA-MB231 cell lines was 1 µM (123, 128, 129, 134, 135). This was therefore chosen for both cell lines. Figure 3.2: Pictorial representation of the changes in the electron transport chain due to oligomycin, FCCP and antimycin-A. Adapted from (136). The second drug injected is FCCP. FCCP collapses the mitochondrial membrane potential by allowing uninhibited proton movement across the mitochondrial membrane (Figure 3.2) (120). Oxygen is maximally consumed by complex IV as complexes in the electron transport chain attempt to maintain the proton gradient across the inner membrane (Figure 3.3: A, C) (136). Cells vary in their sensitivity toward FCCP and therefore the concentration is optimised for maximum response in OCR before cytotoxic levels are reached (137). Mitochondrial stress tests were carried out for both cell lines with a range of FCCP concentrations (0.1 µM, 0.3 µM, 0.5 µM, 0.75 µM, 1 µM, 2 µM). Optimal FCCP concentration was selected for MCF7 as 0.5 µM and for MDA-MB-231 as 1 µM (Figure 3.3: B, D). The last drug injected is Antimycin-A which inhibits complex III, preventing mitochondrial respiration (Figure 3.2) (121). This can be seen as a steep decrease in OCR (Figure 3.3: A, C).Antimycin A-independent oxygen consumption will therefore be due to non-mitochodrial sources. This was optimised based on literature; a concentration of 10 µM was chosen for MCF7 and MDA-MB-231 cells (124). 39 B) 500 400 400 300 300 OCR (%) 500 200 MCF7 OCR (%) A) 200 100 100 0 0.00 0.1 0 50.00 100.00 Time (minutes) 0.5 0.75 150.00 1 0 0.5 1 1.5 2 FCCP concentration (µm) 2.5 0 0.5 2.5 1.5 FCCP (µm) C) D) 800 800 700 600 600 MDA-MB-231 OCR (%) OCR (%) 500 400 400 300 200 200 100 0 0.00 0 50.00 100.00 150.00 Time (minutes) 0.5 0.75 1 1.5 2 FCCP concentration (µm) 1 1.5 FCCP (µm) Figure 3.3: FCCP optimisation of MCF7 and MDA-MB-231 breast cancer cell lines. A) MCF7 OCR graph during FCCP optimisation. B) FCCP optimisation of MCF7 (n=2). C) MDA-MB231 OCR graph during FCCP optimisation. D) FCCP optimisation of MDA-MB-231 (n=2). Data is presented as mean ± SD with OCR measured as % baselined to data point 10 (before FCCP injection). 40 3.1.2 Metabolic characterisation of MCF7 and MDA-MB-231 breast cancer cells The mitochondrial stress test uses injected drugs to knock out components of the electron transport chain, which creates measurable differences that can be attributed to various metabolic parameters. Through sequential injection of oligomycin, FCCP, and antimycin-A the mitochondrial stress test generates measures of basal respiration, ATPlinked respiration, proton leak, maximal respiration, spare capacity, and nonmitochondrial respiration (Figure 3.4). The calculations to determine these values are included in Table 3.1. Figure 3.4: Representation of a typical mitochondrial stress test, identifying the fractions attributed to various metabolic parameters (138). 41 Table 3.1: Calculations to determine metabolic parameters of the mitochondrial stress test (138, 139). Metabolic Parameter Calculation Non-mitochondrial Respiration minimum rate measurement after antimycin-A injection Basal Respiration (last rate measurement before first injection) – (nonmitochondrial respiration) Maximal Respiration (maximum rate measurement after FCCP injection) – (nonmitochondrial respiration) Proton Leak (minimum rate measurement after oligomycin injection) – (non-mitochondrial respiration) ATP-linked Respiration (last rate measurement before oligomycin injection) – (minimum rate measurement after oligomycin injection) Spare Respiratory Capacity (maximal respiration) – (basal respiration) Spare Respiratory Capacity (%) maximal respiration x 100 basal respiration Coupling Efficiency (%) ATP-linked respiration rate x 100 basal respiration rate Baseline OCR last OCR rate measurement before oligomycin injection Baseline ECAR last ECAR rate measurement before oligomycin injection Stressed OCR maximum OCR rate measurement after FCCP injection Stressed ECAR maximum ECAR measurement after oligomycin injection 42 Data for MCF7 was obtained from two biological samples run on two separate XF plates (n=2) and data for MDA-MB-231 was obtained from four biological samples run on two separate XF plates (n=4). There were 6-10 replicate wells per sample. Presented data has been normalised to µg protein (pmol/min/µg protein or mpH/min/µg protein). MCF7 cells had significantly higher basal respiration than MDA-MB-231 cells (1.15 vs 0.77, respectively; p=0.005). MCF7 cells also had higher maximal respiration than MDA-MB-231 cells (2.00 vs 1.51, respectively; p=0.011) (Figure 3.5: A). Assessment of oxygen consumption following inhibition of the ATP synthase with oligomycin indicates that the MCF7 cells are generating higher amounts of ATP than MDA-MB-231 cells (0.88 vs 0.66, respectively; p=0.012). MCF7 cells however, had higher indication of proton leak than MDA-MB-231 cells (0.28 vs 0.11, respectively; p=0.001). In line with this, the coupling efficiency of MDA-MB-231 cells was significantly higher than MCF7 cells (0.85 vs 0.76, respectively; p<0.001) (Figure 3.5: D). Non-mitochondrial respiration did not differ between MCF7 cells and MDA-MB-231 cells (0.18 vs 0.22, respectively; p=0.700). There was no significant difference in spare respiratory capacity between MCF7 and MDA-MB-231 cells (0.84 vs 0.74, respectively; p=0.137) (Figure 3.5: B). Spare capacity can also be measured as a percentage of basal respiration; MDAMB-231 cells had a higher spare capacity as a percentage of basal respiration than MCF7 cells (197 vs 173, respectively; p=0.021) (Figure 3.5: C). 43 A) 2.5 OCR (pmol/min/µg protein) MCF7 MDA-MB-231 2 1.5 1 0.5 0 0.00 20.00 40.00 60.00 80.00 Time (minutes) 100.00 120.00 140.00 B) OCR (pmol/min/µg protein) 2.5 MCF7 MDA-MB-231 * 2 1.5 ** * 1 ** 0.5 0 Basal Respiration * 200 180 160 140 120 100 80 60 40 20 0 Maximal Respiration Spare Respiratory Capacity Non Mito Respiration ATP-linked respiration D) 100 Coupling Efficiency (%) Spare Respiratory Capacity (%) C) Proton Leak ** 80 60 40 20 0 MCF7 MDA-MB-231 MCF7 MDA-MB-231 Figure 3.5: Comparing oxidative metabolic parameters of MCF7 and MDA-MB-231 breast cancer cell lines. A) OCR graph of MCF7 (n=2) and MDA-MB-231 (n=4). B) Various oxidative metabolic parameters of MCF7 (n=2) and MDA-MB-231 (n=4). C) Spare respiratory capacity as a percentage of basal respiration between MCF7 (n=2) and MDA-MB-231 (n=4). Basal respiration indicated by dashed line at 100%. D) Coupling efficiency of MCF7 (n=2) and MDA-MB-231 (n=4). All data normalised to protein content and presented as mean ± SD as analysed by independent twosample t-test. Significance denoted as * p<0.05, ** p<0.01, *** p<0.001. 44 The data from a mitochondrial stress test can also be displayed as an energy graph (Figure 3.6). This is generated by the Seahorse Energy Phenotype report generator from measurements of baseline and stressed OCR and ECAR. Baseline is used to refer to the OCR or ECAR before any drug injection. Stressed is used to refer to OCR directly after FCCP injection and ECAR directly after oligomycin injection. The graph is designed to inform the general energy phenotype of the cells being tested Figure 3.6: Energy graph displaying cells in their baseline and stressed phenotypes. Obtained from (139). 45 MCF7 cells showed a significant increase in OCR between baseline and stressed conditions (1.34 vs 2.18, respectively; p=0.005). MDA-MB-231 cells also had higher OCR in stressed conditions than baseline conditions (1.73 vs 0.99, respectively; p<0.001). Baseline OCR rate was higher in MCF7 cells than MDA-MB-231 cells (1.34 vs 0.99, respectively; p=0.016). MCF7 cells also had a higher stressed OCR rate compared to MDA-MB-231 (2.18 vs1.73, respectively; p=0.012) (Figure 3.7: A). MCF7 cells showed no difference in ECAR rate between baseline and stressed conditions (0.10 vs 0.18, respectively; p=0.153); although an upward trend was observed. In comparison MDA-MB-231 cells had a significant increase in ECAR rate between baseline and stressed conditions (0.06 vs 0.11, respectively; p<0.001). There was no difference in baseline ECAR rate between MCF7 and MDA-MB-231 (0.10 vs 0.06, respectively; p=0.281). MCF7 cells had a higher stressed ECAR rate than MDA-MB231 (0.18 vs 0.11, respectively; p=0.039) (Figure 3.7: B). When this data is visualised on the energy graph both cell lines showed a similar energy phenotype; MDA-MB-231 cells move from quiescent to aerobic and MCF7 cells move from a low level of aerobic to highly aerobic. Both cell lines remained quiescent in terms of glycolytic ability (Figure 3.7: C). 46 B) A) ** 2.5 0.25 ** ** * 2 1.5 ECAR (mpH/min/µg protein) OCR (pmol/min/µg protein) * * 1 0.5 0 0.2 0.15 0.1 0.05 0 Baseline Stressed Baseline MCF7 Stressed MDA-MB-231 C) 2.5 aerobic MDA-MB-231 energetic MCF7 Baseline 2 OCR (pmol/min/µg protein) Stressed 1.5 1 0.5 quiescent glycolytic 0 0 0.2 0.4 0.6 ECAR (mpH/min/µg protein) 0.8 1 Figure 3.7: Comparing energy phenotypes of MCF7 and MDA-MB-231 breast cancer cell lines using OCR and ECAR in baseline and stressed conditions. A) OCR in baseline and stressed conditions for MCF7 (n=2) and MDA-MB-231 (n=4). B) ECAR in baseline and stressed conditions for MCF7 (n=2) and MDA-MB-231 (n=4). C) Energy graph of MCF7 and MDA-MB-231. All data normalised to protein content and presented as mean ± SD as analysed by independent two-sample t-test. Significance denoted as * p<0.05, ** p<0.01, *** p<0.001. 47 3.1.5 Metabolic characterisation of MCF7 breast cancer cells grown alone and in adipocyte co-culture Data for MCF7 was obtained from two biological samples run on two separate XF plates (n=2) and data for MCF7 co-cultured with adipocytes was obtained from two biological samples run on two separate XF plates (n=2). There were 6-10 replicate wells per sample. Presented data has been normalised to µg protein. There was no significant difference between MCF7 cells grown alone or in adipocyte coculture for any of the oxidative metabolic parameters (Figure 3.8: B). The MCF7 cells co-cultured with adipocytes tended towards a lower OCR for all of the parameters measured (Figure 3.8: A). Basal respiration was unchanged between MCF7 cells and cocultured MCF7 (1.10 vs 0.98, respectively; p=0.379), as was maximal respiration (1.91 vs 1.55, respectively; p=0.322). MCF7 and co-cultured MCF7 cells did not differ in proton leak (0.25 vs 0.17, respectively; p=0.172). They also had no difference in ATPlinked respiration (0.85 vs 0.81, respectively; p=0.453). Non-mitochondrial respiration was unaltered between MCF7 and co-cultured MCF7 (0.22 vs 0.15, respectively; p=0.169). MCF7 and co-cultured MCF7 showed no change in spare respiratory capacity (0.81 vs 0.56, respectively; p=0.273), or spare capacity measured as percentage of basal respiration (174 vs 157, respectively: p=0.245) (Figure 3: C). There was no difference in coupling efficiency between MCF7 and co-cultured MCF7 cells (0.77 vs 0.83, respectively; p=0.661) (Figure 3.8: D). 48 A) 2.5 ECAR (mpH/min/µg protein) MCF7 MCF7 co-culture 2 1.5 1 0.5 0 0.00 20.00 40.00 60.00 80.00 Time (minutes) 100.00 120.00 140.00 B) OCR (pmol/min/µg protein) 2.5 MCF7 MCF7 co-culture 2 1.5 1 0.5 0 Basal Respiration Proton Leak Spare Respiratory Capacity Non Mito Respiration ATP-linked respiration D) 200 100 Coupling Efficiency (%) Spare Respiratory Capacity (%) C) Maximal Respiration 160 120 80 40 80 60 40 20 0 0 MCF7 MCF7 coculture MCF7 MCF7 coculture Figure 3.8: Comparing oxidative metabolic parameters of MCF7 and MCF7 co-cultured with adipocytes. A) OCR graph of MCF7 (n=2) and co-cultured MCF7 (n=2). B) Various oxidative metabolic parameters of MCF7 (n=2) and co-cultured MCF7 (n=2). C) Spare respiratory capacity as a percentage of basal respiration between MCF7 (n=2) and co-cultured MCF7 (n=2). Basal respiration indicated by dashed line at 100%. D) Coupling efficiency of MCF7 (n=2) and co-cultured MCF7 (n=2). All data normalised to protein content and presented as mean ± SD as analysed by independent twosample t-test. 49 MCF7 cells showed an increase in OCR between baseline and stressed conditions (1.34 vs 2.18, respectively; p=0.005), whereas co-cultured MCF7 cells showed no change (0.89 vs 1.28, respectively; p=0.505). There was no difference in baseline OCR between MCF7 and co-cultured MCF7 cells (1.34 vs 0.89, respectively; p=0.340). Stressed OCR also showed no difference between MCF7 and co-cultured MCF7 cells (2.18 vs 1.28, respectively; p=0.307). Although no significant difference was observed, there was a noticeable trend toward decreased OCR in co-cultured MCF7 cells compared to MCF7 cells grown alone (Figure 3.9: A). Co-cultured MCF7 cells had a significant increase in ECAR between baseline and stressed conditions (0.04 vs 0.10, respectively; p=0.012), whereas MCF7 cells grown alone showed no change in ECAR between baseline and stressed conditions (0.10 vs 0.18, respectively; p=0.153) (Figure 3.9: B). Adipocyte co-cultured MCF7 cells showed greater variation in OCR than in ECAR whilst MCF7 cells grown alone showed greater variation in ECAR than OCR. MCF7 cells cultured alone had a more energetic phenotype than their adipocyte cocultured counterparts; starting in a low level aerobic state and moving to a high level aerobic state whilst the co-cultured cells started in a quiescent state and moved into a low level aerobic state (Figure 3.9: C). 50 B) A) 2.5 0.25 ** ECAR (mpH/min/µg protein) OCR (pmol/min/µg protein) * 2 1.5 1 0.5 0 0.2 0.15 0.1 0.05 0 Baseline Stressed Baseline MCF7 Stressed MCF7 co-culture C) 2.5 aerobic MCF7 energetic MCF7 co-culture OCR (pmol/min/µg protein) Baseline 2 Stressed 1.5 1 0.5 glycolytic quiescent 0 0 0.2 0.4 0.6 0.8 1 ECAR (mpH/min/µg protein) Figure 3.9: Comparing energy phenotypes of MCF7 and adipocyte co-cultured MCF7 using OCR and ECAR in baseline and stressed conditions. A) OCR in baseline and stressed conditions for MCF7 (n=2) and co-cultured MCF7 (n=2). B) ECAR in baseline and stressed conditions for MCF7 (n=2) and co-cultured MCF7 (n=2). C) Energy graph of MCF7 and co-cultured MCF7. All data normalised to protein content and presented as mean ± SD as analysed by independent two-sample t-test. Significance denoted as * p<0.05, ** p<0.01, *** p<0.001. 51 3.1.6 Fatty acid oxidation stress test optimisation The fatty acid oxidation stress test began optimisation as a protocol for measuring alteration in β-oxidation when breast cancer cells were co-cultured with adipocytes. This stress test is capable of calculating the respiration attributable to either exogenous or endogenous fatty acids (131). This is by addition of a free fatty acid, palmitate (SigmaAldrich), to half of the wells via a protein vehicle, bovine serum albumin (BSA). BSA is required for cells to be able to utilise the palmitate. A BSA control is applied to the other half of the wells. On top of this, the CPT1 inhibitor, etomoxir (Sigma-Aldrich), is applied to half the wells from each of the BSA and BSA:palmitate conditions. Etomoxir prevents fatty acid oxidation by stopping palmitate from entering the mitochondria (131). This gives four well conditions which can be used to calculate the metabolic parameters of fatty acid oxidation (Figure 3.10). The calculations required to work out the final values can be seen in Table 3.2. Figure 3.10: Representation of a typical fatty acid oxidation stress test, identifying the fractions attributed to exogenous or endogenous oxidation (140). 52 Table 3.2: Calculations to determine metabolic parameters of the fatty acid oxidation stress test (131). Metabolic Parameter Calculation Oxygen Consumption due to uncoupling by free fatty acids (oligomycin BSA:palmitate –etomoxir rate) – (oligomycin BSA –etomoxir rate) Basal Respiration due to utilization of exogenous fatty acids (basal BSA:palmitate –etomoxir rate) – (basal BSA –etomoxir rate) – (OCR due to uncoupling by free fatty acids) Maximal Respiration due to utilization of exogenous fatty acids (maximal BSA:palmitate –etomoxir rate) – (maximal BSA – etomoxir rate) – (OCR due to uncoupling by free fatty acids) Basal Respiration due to utilization of endogenous fatty acids (basal BSA –etomoxir rate) – (basal BSA +etomoxir rate) Maximal Respiration due to utilization of endogenous fatty acids (maximal BSA –etomoxir rate) – (maximal BSA +etomoxir rate) 3.1.6.1 BSA:palmitate conjugation A key component for a FAO stress test is the BSA:palmitate conjugate which allows the fatty acid, palmitate, to be taken up by the cell via the vehicle, BSA. These can be conjugated together through a series of heating steps (132). Multiple attempts were made to conjugate the BSA to palmitate. In the first attempt, the instructions were correctly followed at a 1/10 scaled down version. Palmitate required dissolving and gradually heating to 70°C at which point the solution is supposed to transition from cloudy to clear. The solution did not reach a clear state and was not continued with. Further attempts were made, heating the palmitate solution to 80-100°C. The solution remained cloudy at higher temperatures but did appear to be slightly less cloudy than at 70°C. These attempts were then repeated with a 1/5 scaled down version and a full size version with no apparent changes observed. Conjugation was eventually carried out using the original protocol, with the observation that there had not been a state change from cloudy to clear noted. It was assumed that the conjugation was successful. 3.1.6.2 Optimisation of FCCP for fatty acid oxidation stress test Cell lines that have undergone FCCP optimisation for the mitochondrial stress test have to repeat this in the presence of BSA. This is because BSA can bind FCCP and so greater concentrations are often required (131). FCCP optimisation in the presence of BSA was very variable and no concentration could be accurately selected (Figure 3.11). OCR values were at similar levels for both BSA and BSA:palmitate optimisations (Figure 3.12). 53 350 300 OCR (%) 250 200 150 100 50 0 0 1 2 3 4 5 6 FCCP concentration (µM) Figure 3.11: FCCP optimisation of MCF7 cells in the presence of BSA. Data was obtained from two separate experiments and combined to cover the range presented. 350.0 Palmitate:BSA BSA OCR (pmol/min/µg protein) 300.0 250.0 200.0 150.0 100.0 50.0 0.0 0.00 20.00 40.00 60.00 Time (minutes) 80.00 100.00 120.00 Figure 3.12: MCF7 OCR graph during FCCP optimisation with BSA or BSA:palmitate. Data presented as mean ± SD from 10 wells. 54 3.2 Analysis of proteins involved in metabolism using Western blotting Western blotting was used to investigate levels of CPT1A, ACC1, phospho-ACC and GPD2 in MCF7 and MDA-MB-231 cells and their adipocyte co-cultured counterparts. CPT1A is located in the mitochondrial membrane and is involved in the translocation of fatty acids from the cytoplasm into the mitochondrial matrix where they can be used in β-oxidation (105). ACC catalyses the conversion of acetyl CoA into malonyl CoA which is an inhibitor of CPT1A. The phosphorylated form of ACC is inactive and does not catalyse this reaction (95). GPD2 is a protein in the glycerophosphate shuttle that has been associated with changes in cellular metabolism and upregulation in prostate cancer (110, 111). Data is calculated as relative density to β-actin. 3.2.1 CPT1A CPT1A was detected in both cell lines at 88kDa (Figure 3.13: A). MCF7 cells had significantly more CPT1A protein than MDA-MB-231 cells (1.11 vs 0.46, respectively; p<0.001) (Figure 3.13: B). There was no difference in CPT1A protein level between cocultured MCF7 and MCF7 grown alone (0.64 vs 0.66, respectively; p=0.538) (Figure 3.13: C, D). There was also no significant difference in CPT1A between co-cultured MDA-MB-231 and MDA-MB-231 grown alone (0.60 vs 0.57, respectively; p=0.768) (Figure 3.13: C, E). 55 B) MCF7 A) 1 Relative Density to β-actin 1.4 MDA-MB-231 2 3 1 2 3 CPT1A β-actin 88kDa 64kDa 42kDa 37kDa *** 1.2 1 0.8 0.6 0.4 0.2 0 MCF7 MDA-MB-231 C) CPT1A MCF7 64kDa β-actin 37kDa β-actin 37kDa MCF7 0.8 E) MDA-MB-231 0.9 0.8 0.7 Relative Density to β-actin Relative Density to β-actin D) MDA-MB-231 CPT1A 64kDa 0.6 0.5 0.4 0.3 0.2 0.1 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 Control Co-culture Control Co-culture Figure 3.13: Comparison of CPT1A protein level in MCF7 and MDA-MB-231 cells compared to their adipocyte co-cultured counterparts. A) Western blot comparison of CPT1A in MCF7 and MDA-MB-231. B) CPT1A in MCF7 (n=3) and MDA-MB-231 (n=3). C) Western blot comparison of CPT1A in MCF7 and co-cultured MCF7 and in MDA-MB-231 and co-cultured MDA-MB-231. D) CPT1A MCF7 (n=5) and co-cultured MCF7 (n=6). E) CPT1A in MDA-MB-231 (n=5) and co-cultured MDA-MB-231 (n=6). Data normalised to β-actin and presented as mean ± SD as analysed by independent two-sample t-test. Significance denoted as * p<0.05, ** p<0.01, *** p<0.001. 56 2.3 ACC1 and phospho-ACC ACC1 was detected in both cell lines at 265kDa (Figure 3.14: A). There was no difference in ACC1 level between MCF7 cells and MDA-MB-231 cells (0.63 vs 0.65, respectively; p=0.889) (Figure 3.14: B). There was no difference in ACC1 protein levels between co-cultured MCF7 and MCF7 grown alone (0.50 vs 0.69, respectively; p=0.061); although there was an observed trend downward in co-cultured MCF7 compared to MCF7 grown alone (Figure 3.14: C, D). ACC1 protein level did not differ between co-cultured MDA-MB-231 and MDA-MB-231 grown alone (0.97 vs 1.05, respectively; p=0.380) (Figure 3.14: C, E). Phospho-ACC was detected in both cell lines at 285kDa (Figure 3.15: A). Phospho-ACC level was not different between MCF7 cells and MDA-MB-231 cells (0.89 vs 0.90, respectively; p=0.866) (Figure 3.15: B). There was a significant increase in phosphoACC in co-cultured MCF7 compared to MCF7 grown alone (1.03 vs 0.50, respectively; p=0.013) (Figure 3.15: C, D). Phospho-ACC was also significantly increased in cocultured MDA-MB-231 compared to MDA-MB-231 grown alone (1.11 vs 0.56, respectively; p=0.007) (Figure 3.15: C, E). 57 A) MCF7 MDA-MB-231 1 2 1 2 ACC1 265kDa 250kDa β-actin 50kDa 42kDa Relative Density to β-actin B) 1 0.8 0.6 0.4 0.2 0 MCF7 MDA-MB-231 C) ACC1 250kDa MCF7 β-actin 39kDa β-actin MDA-MB-231 ACC1 250kDa 39kDa D) MCF7 0.8 E) 1.4 Relative Density to β-actin Relative Density to β-actin 0.9 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 MDA-MB-231 1.2 1 0.8 0.6 0.4 0.2 0 Control Co-culture Control Co-culture Figure 3.14: Comparison of ACC1 protein level in MCF7 and MDA-MB-231 cells compared to their adipocyte co-cultured counterparts. A) Western blot comparison of ACC1 in MCF7 and MDA-MB-231. B) ACC1 in MCF7 (n=3) and MDAMB-231 (n=4). C) Western blot comparison of ACC1 in MCF7 and co-cultured MCF7 and in MDA-MB231 and co-cultured MDA-MB-231. D) ACC1 MCF7 (n=5) and co-cultured MCF7 (n=6). E) ACC1 in MDA-MB-231 (n=5) and co-cultured MDA-MB-231 (n=6). Data normalised to β-actin and presented as mean ± SD as analysed by independent two-sample t-test. 58 B) 1 MDA-MB-231 2 1 2 P-ACC 280kDa 250kDa β-actin 50kDa 42kDa Relative Density to β-actin 1.2 MCF7 A) 1 0.8 0.6 0.4 0.2 0 MCF7 MDA-MB-231 C) MCF7 P-ACC 250kDa β-actin 39kDa β-actin MDA-MB-231 P-ACC 250kDa 39kDa D) MCF7 E) 1.4 * MDA-MB-231 ** 1.4 Relative Density to β-actin Relative Density to β-actin 1.6 1.2 1 0.8 0.6 0.4 0.2 0 1.2 1 0.8 0.6 0.4 0.2 0 Control Co-culture Control Co-culture Figure 3.15: Comparison of phospho-ACC protein level in MCF7 and MDA-MB-231 cells compared to their adipocyte co-cultured counterparts. A) Western blot comparison of phospho-ACC in MCF7 and MDA-MB-231. B) Phospho-ACC in MCF7 (n=4) and MDA-MB-231 (n=3). C) Western blot comparison of phospho-ACC in MCF7 and co-cultured MCF7 and in MDA-MB-231 and co-cultured MDA-MB-231. D) Phospho-ACC MCF7 (n=5) and cocultured MCF7 (n=6). E) Phospho-ACC in MDA-MB-231 (n=5) and co-cultured MDA-MB-231 (n=6). Data normalised to β-actin and presented as mean ± SD as analysed by independent two-sample t-test. Significance denoted as * p<0.05, ** p<0.01, *** p<0.001. 59 3.2.4 GPD2 GPD2 was detected in both cell lines. A double band was observed in MCF7 cells at 81kDa and 68kDa. In MDA-MB-231 cells GPD2 was only visualised as a single band at 68kDa (Figure 3.16: A). There was no difference in GPD2 protein level between MCF7 and MDA-MB-231 (1.13 vs 0.66, respectively; p=0.071) (Figure 3.16: B). There was no difference in overall GPD2 protein level between MCF7 grown alone and co-cultured MCF7 (0.70 vs 0.57, respectively; p=0.533) (Figure 3.17: A, B). GPD2 81kDa protein increased in MCF7 grown alone compared to co-cultured MCF7 (0.21 vs 0.09, respectively; p=0.013) (Figure 3.17: D). However, GPD2 68kDa protein was not significantly different between MCF7 grown alone and co-cultured MCF7 (0.15 vs 0.29, respectively; p=0.127); although there was a trend toward increased protein in the cocultured MCF7 samples compared to MCF7 grown alone (Figure 3.17: E). For MDAMB-231 there was no difference in GPD2 68kDa protein between cells grown alone and co-cultured cells (0.20 vs 0.01, respectively; p=0.069); although protein in the cocultured MDA-MB-231 samples had a decreasing trend compared to MDA-MB-231 grown alone (Figure 3.17: A, C). B) MCF7 A) 1 42kDa 37kDa 3 1 2 3 GPD2 β-actin 81kDa 68kDa 2 MDA-MB-231 Relative Density to β-actin 1.6 1.2 0.8 0.4 0 MCF7 MDA-MB-231 Figure 3.16: Comparison of GPD2 protein level between MCF7 and MDA-MB-231. A) Western blot comparison of GPD2 in MCF7 and MDA-MB-231. B) GPD2 in MCF& (n=3) and MDAMB-231 (n=3). Data normalised to β-actin and presented as mean ± SD as analysed by independent two-sample t-test. 60 A) GPD2 64kDa β-actin 37kDa Relative Density to β-actin 1.2 MCF7 total C) 0.4 Relative Density to β-actin B) 1 0.8 0.6 0.4 0.2 Control 0.3 0.2 0.1 Co-culture MCF7 81kDa Control E) 0.5 ** Relative Density to β-actin Relative Density to β-actin 0.3 MDA-MB-231 0 0 D) MDA-MB-231 GPD2 64kDa MCF7 β-actin 37kDa 0.25 0.2 0.15 0.1 0.05 Co-culture MCF7 68kDa 0.4 0.3 0.2 0.1 0 0 Control Co-culture Control Co-culture Figure 3.17: Comparison of GPD2 protein level in MCF7 and MDA-MB-231 cells compared to their adipocyte co-cultured counterparts. A) Western blot comparison of GPD2 in MCF7 and co-cultured MCF7 and in MDA-MB-231 and cocultured MDA-MB-231. B) Total GPD2 in MCF7 (n=5) and co-cultured MCF7 (n=6). C) 81kDa GPD2 in MCF7 (n=5) and co-cultured MCF7 (n=6). D) 68kDa GPD2 in MCF7 (n=5) and co-cultured MCF7 (n=6). E) 81kDa GPD2 in MDA-MB-231 (n=5) and co-cultured MDA-MB-231 (n=6). Data normalised to β-actin and presented as mean ± SD as analysed by independent two-sample t-test. Significance denoted as * p<0.05, ** p<0.01, *** p<0.001. 61 3.3 Analysis of glycerol in conditioned media There was no difference in glycerol level between the conditioned media conditions for MCF7 (ANOVA (3,22)=0.42, p=0.744). There was also no difference for MDA-MB-231 (ANOVA (3,14)=0.44, p=0.727). Overall there was no significant difference in glycerol levels between MCF7 and MDA-MB-231 for any of the conditioned media conditions (ANOVA (7,36)=0.37, p=0.912) (Figure 3.18). When glycerol levels for post co-culture samples and 24 hours post co-culture samples were combined and compared between MCF7 and MDA-MB-231 there was no difference (34 µM vs 44 µM, respectively; p=0.077). It was observed that co-cultured samples from different biological adipocyte sources showed differing patterns of glycerol levels across the conditioned media conditions and between breast cancer cell lines (Supplementary Figure 5.5). 140 MCF7 MDA-MB-231 120 Glycerol (µM) 100 80 60 40 20 0 Pre-adipocyte Mature adipocyte Post co-culture 24hr post co-culture Figure 3.18: Comparison of glycerol levels in various conditioned media samples from MDA-MB-231 and MCF7 adipocyte co-cultures. A) Glycerol in MCF7 (n=7) and MDA-MB-231 (n=3) co-culture conditioned media. B) Glycerol in MCF7 and MDA-MB-231 co-culture conditioned media from seven separate adipocyte samples. Data presented as mean ± SD as analysed by one-way ANOVA. 62 4 Discussion In this project it was hypothesised that breast tumour cells are capable of activating lipid release by nearby adipocytes, and may metabolise the resulting glycerol and fatty acids to fuel migration and invasion. This was based on recent literature which has found that breast cancer cell interactions with adipocytes causes formation of cancer-associated adipocytes (CAA), which result in more aggressive and invasive breast cancer and resistance to radiotherapy and chemotherapy (15-17). Additionally to this, adipocytes have been observed to transfer lipids to neighbouring ovarian cancer cells where they are metabolised by β-oxidation (17). The overall aim was to observe metabolism in breast tumour cells and look at potential changes when the breast tumour cells are co-cultured with adipocytes. 4.1 Analysis of metabolism in breast cancer cells and breast cancer cells cocultured with adipocytes using the Seahorse Extracellular Flux Analyser The objective of these experiments was to characterise tumour cell metabolism in breast cancer cell lines MCF7 and MDA-MB-231. Further to this, tumour cell metabolism would be investigated in MCF7 cells co-cultured with adipocytes to see if breast cancer cells are altering metabolism; potentially utilising glycerol and fatty acids from adipocyte lipolysis. Firstly, it was hypothesised that MDA-MB-231 cells would be more metabolically active than MCF7 cells due to their more metastatic phenotype. Secondly, that MCF7 cells grown in co-culture would be more metabolically active than MCF7 cells grown alone, as they are metabolising glycerol and free fatty acids. Optimisation of the Seahorse protocols to the chosen cell lines was a very important part of the project. If optimisation steps are not completed then the data gathered may not show the true metabolic behaviour of the cells and cannot be validly interpreted. The decision to not use Cell-Tak, although it showed good adhesion properties, was based on observed abnormal cell morphology observed in MCF7 and MDA-MB-231 cells when cultured on Cell-Tak. It was thought that, therefore, their phenotype might be altered and experimental results would not be valid. At the optimised seeding density cells looked morphologically similar to when they were grown in a culture flask. They also behaved within the sensitivity parameters specified by Seahorse Biosciences which would aid accuracy of readings and reliability of data from the ensuing experiments. This density was further supported with an MTT assay which ensured that cell viability was not being 63 reduced at this density. Seeding conducted in co-cultured cells had to be altered between experiments and a final seeding density of 6x104 was confirmed by the third co-culture experiment. Unfortunately this meant that the seeding density varied across co-culture experiments and 85% confluence was only achieved in the last experiment. It was generally observed that MCF7 cells were more metabolically active than MDAMB-231 cells. They had higher basal and maximal respiration, and higher ATP-linked respiration. This was an unexpected result and did not support the initial hypothesis. MDA-MB-231 cells are typically described as a more migratory breast cancer cell line (119, 141). As such, it was predicted that they would be more metabolically active than MCF7 cells. Although MDA-MB-231 cells had lower respiration, they had a higher spare capacity as a percentage of their basal respiration; almost doubling their respiration levels. This showed the superior ability of MDA-MB-231 cells to increase their OCR in response to stressful conditions compared to MCF7 cells. It has been proposed that vitality and survival of cells may be dependent on the maintenance of bioenergetic capacity during conditions of maximal physiological stress (121). This is an important survival advantage for MDA-MB-231 (74-76). MCF7 cells had a higher proton leak than MDA-MB-231 cells, and in line with this, MDA-MB-231 cells had a higher coupling efficiency. Coupling is the mechanism by which ATP is generated (142). This is by the proton gradient being coupled to ATP synthesis through ATP synthase (142). This means that MDA-MB-231 cells were more efficient at synthesising ATP. This is important as cancer cells need to maximise resources and this feature may give MDA-MB-231 cells an advantage over MCF7 cells. In combination with the ability to greatly increase OCR under stress, MDA-MB-231 cells have two important survival advantages over MCF7 cells. Proton leak is suggested to be a mechanism that helps prevent reactive oxygen species (ROS) formation (143). The presence of ROS is part of a fine balance; increased ROS can promote tumour development, but high levels of ROS can result in cell death (87, 88). It is possible that the low proton leak in MDA-MB-231 cells has led to higher levels of ROS and that this is indicative of the aggressive nature of these cells. This is supported by research from Pelicano et al. (2014), who have observed increased ROS in triple negative breast cancer cell lines (144). 64 When looking at the energy graph, in baseline (non-metabolically stressed) conditions, MDA-MB-231 cells were displayed as quiescent and MCF7 cells as low level aerobic. Once stressed by the drug injections the cells moved into mid-high level aerobic state; their glycolysis however remained quiescent. The Seahorse generated energy phenotype depicts both MCF7 and MDA-MB-231 as non-glycolytic, whether in baseline or stressed conditions. This infers that both cell lines are not subject to the modern interpretation of Warburg’s Theory, which details that in aerobic conditions cells will sustain oxidative phosphorylation and become glycolytic (76, 80). However, as described below, the MDA-MB-231 cells did show a significant increase in ECAR, a measure of glycolysis, between baseline and stressed conditions. The energy graph generated by the Seahorse Energy Phenotype Report Generator provides no explanation for the quadrant divisions and overall scale. It is probable that the cells originally used to develop these quadrants created divisions that are not applicable to a wide range of cell types. This would mean that the definition of MCF7 and MDA-MB-231 as quiescent, energetic, aerobic, or glycolytic is somewhat arbitrary. Therefore, observations of aerobic and glycolytic changes should be interpreted from changes in cellular OCR and ECAR, not from where the cells are depicted on the Seahorse generated energy graph. Both cell lines increased in oxidative respiration between baseline and stressed conditions, but MCF7 cells, in line with other calculations of oxidative respiration, had significantly higher OCR rates than MDA-MB-231 in both conditions. MDA-MB-231 cells, however, had an increase in glycolysis when stressed, whilst MCF7 cells showed no change. This upregulation of glycolysis in MDA-MB-231 cells is in agreement with Warburg’s theory, and correlates with MDA-MB-231 being considered a more aggressive cell line. This is also supported by Gupta and Tikoo (2013), who have observed increased viability and proliferation in MDA-MB-231 cells, but not MCF7 cells, exposed to high glucose levels (145). This may be due to the ability of MDA-MB231 to increase glycolysis and take advantage of this fuel. MCF7 cells did, conversely, have a higher ECAR than MDA-MB-231 cells in metabolically stressed conditions, but were unchanged between baseline and stressed conditions. This was seen in oxidative spare capacity as well and shows a greater ability of MDA-MB-231, compared to MCF7, to adapt under stressed conditions. This adaptive ability is beneficial for cancerous growth as it allows for survival in changing 65 environments that may stress the cells (74-76). As mentioned previously, maintaining bioenergetic capacity may be important for vitality and survival of cells (121). This is supported by the literature which describes MDA-MB-231 and other triple negative cell lines as more aggressive than MCF7 and other luminal cell lines (119, 141). Previous literature using the Seahorse Extracellular Flux Analyser to look at MCF7 and MDA-MB-231 cell lines did not compare the two and data is difficult to compare across papers due to differences in protocols and normalisation procedures (134, 146, 147). This characterisation of MCF7 and MDA-MB-231 cell lines in our lab is useful for future work. Environmental changes can now be introduced, and alternative stress tests run to see how cellular metabolism reacts. It was observed that MCF7 cells grown in adipocyte co-culture were not significantly different from cells grown alone, for most standard measures of oxidative respiration. The only difference seen was an increase in OCR from baseline to stressed conditions in MCF7 cells grown alone but not in co-cultured cells. However, the MCF7 cells cocultured with adipocytes had a significant increase in glycolysis from baseline to stressed conditions, not seen in cells grown alone. This partially supported the hypothesis, as it was anticipated that metabolism would increase in co-cultured MCF7 cells; glycolysis was observed to be upregulated but oxidative respiration was not. During co-cultures, lipid loss and morphological changes, associated with transition of adipocytes to cancer-associated adipocytes (CAA), were visually observed (Supplementary Figure 5.1). It was predicted that fatty acids were being released by CAA into the media where they would become available to the breast cancer cells. It was expected that growing MCF7 with CAA would result in observations of increased oxidative respiration as cells took advantage of the available fatty acids. Nieman and colleagues (2011) have seen transfer, and subsequent β-oxidation, of fatty acids in an analogous ovarian cancer model (17). This metabolic research does not yet support this in a breast cancer model. It is probable that changes in β-oxidation would not be visible in a standard mitochondrial stress test and that fatty acid oxidation stress tests will be necessary to determine if this is occurring. Fatty acids are not provided during mitochondrial stress tests. Therefore, if the β-oxidation pathway was upregulated by adipocyte co-culture the MCF7 cells may not be able to show any oxidative advantages 66 post co-culture (during the mitochondrial stress test) if there is no substrate available for this pathway. Although unexpected, the changes observed in co-culture glycolysis are supported by the Warburg effect (76, 80). When CAA undergo lipolysis they release not only fatty acids but also glycerol. Glycerol can be metabolised through the glycolysis pathway (148). Glycolysis may be upregulated in co-cultured MCF7 cells in response to this glycerol availability. This is also supported by studies of two-compartment tumour metabolism, which hypothesise the release of metabolites from stromal cells being utilised by cancer cells (19, 46, 63). Unlike the β-oxidation pathway, the glycolytic pathway would still have fuel available during the mitochondrial stress test as the assay media is supplemented with glucose. This would allow the observation of glycolysis upregulation that potentially couldn’t be made in β-oxidation. The analysis of energy phenotype in co-cultured MCF7 cells detailed the cells as quiescent to low level aerobic. As mentioned in the comparison of MCF7 and MDAMB-231 cells, this energy phenotype graph is not valid for this research. The aim is to see changes in comparison to the control sample, not in comparison to the cell types that were used to design the energy phenotype graph. In addition to this, the co-cultured MCF7 cells are grown in serum-free conditions as FBS can affect the adipocytes. This means that the cells will not be operating at their optimum performance. An issue with the interpretation of co-culture results is the MCF7 control sample. The controls used for comparison in this project were MCF7 cells grown in standard culture conditions. The co-culture cells, in comparison, were grown on inserts and in serum free media for 3 days. A relevant MCF7 control sample would be grown in identical conditions. This was intended, but not completed, due to time restrictions and equipment issues. Therefore, it must be kept in mind that all MCF7 co-culture data was compared to MCF7 cells grown in standard conditions., Although, previous research in our lab group observed little change in viability, migration rates, or resistance to chemotherapy agents between MCF7 cells grown on inserts and in serum free media compared to those grown in standard co-culture conditions (Supplementary Figure 5.6), it is still possible that metabolism could be affected. However, baseline readings conducted in this study between the measured MCF7 control and a serum free media MCF7 control showed no difference in baseline metabolism. During attempts to grow MCF7 controls in serum free 67 media for 3 days, high levels of cell death were observed and there was difficulty in obtaining sufficient cells. It is promising that, although adipocyte co-cultured MCF7 cells are also in serum free media for 3 days, they show increased viability compared to MCF7 cells cultured in serum free media. This suggests a protective effect of adipocytes in the co-culture which may be partly due to metabolic changes. Final sample sizes for Seahorse experiments were small; 2-4 biological replicates per condition. This was due to time restraints, equipment issues, and restricted availability of adipose tissue samples for co-culture. Adipose samples are very variable and difficult to differentiate to a workable mature state. Because of this, the low intake of adipose samples can be further restricted by sample loss. A Seahorse 24-well plate provides 20 active wells for measurements. The breast cancer cell lines, MCF7 and MDA-MB-231, had reliable results with low deviation within and between plates. For this reason, it is thought that the data obtained from the Seahorse experiments, although from small sample sizes, is representative of these breast cancer cell lines. It must also be taken into account that breast cancer cell lines were only investigated within a narrow passage range. Optimisation of the Seahorse fatty acid oxidation test was started during this project. This test will allow analysis of changes in β-oxidation between breast cancer cells grown alone or in adipocyte co-culture. Issues in the BSA:palmitate conjugation slowed progress and consequently, optimisation is still ongoing. It was not apparent that BSA:palmitate conjugation had been successful. FCCP optimisation for MCF7 was also incomplete. During FCCP optimisation it was observed that when MCF7 cells were exposed to BSA or BSA:palmitate the OCR after FCCP addition appeared unchanged. This was unexpected as it was anticipated that the MCF7 cells would have increased OCR in the BSA:palmitate condition as this was depicted in the protocol example (131). It is possible that MCF7 cells do not have active β-oxidation in baseline conditions and so when supplied with palmitate cannot take advantage of it. However, it is equally possible that the BSA:palmitate conjugation was unsuccessful and that the palmitate was therefore not available for the cells during the stress test. To check the BSA:palmitate conjugation two experiments were proposed: a fatty acid oxidation test on HepG2, the cell line shown in the protocol; and to run the BSA and the BSA:palmitate on a gel to see if any differences are observed. Neither of these were carried out in this project due to time constraints. 68 4.2 Analysis of proteins involved in β-oxidation using Western blotting The objective of these experiments was to investigate if β-oxidation of fatty acids is upregulated in adipocyte co-culture. This was analysed by determining the levels of specific tumour cell cytosolic and mitochondrial proteins; CPT1A (generates acylcarnitine to aid fatty acid transfer into the mitochondrial matrix), ACC1 (catalyses conversion of acetyl-CoA to malonyl-CoA), and phosphorylated ACC (inactive ACC) (86, 95, 96). Additionally, GPD2 was looked at, as some studies have shown associations between GPD2 and changes in cancer cellular metabolism (110, 111, 114). It was hypothesised that CPT1A levels would increase in cells co-cultured with adipocytes compared to those grown alone. Secondly, that ACC1 levels would not change, but phosphorylated ACC levels would increase in adipocyte co-cultured cells compared to cells grown alone. MCF7 cells showed a higher level of CPT1A than MDA-MB-231 cells. This is supported by Kim (2015), who also observed increased CPT1A in MCF7 cells compared to MDAMB-231 cells (149). Higher protein levels of CPT1A have been shown to result in increased β-oxidation (150, 151). This suggests that MCF7 cells have an increased ability to oxidise fatty acids compared to MDA-MB-231 cells. There was no difference in CPT1A protein level between co-culture with adipocytes, or cells grown alone, for MCF7 or MDA-MB-231. This was unexpected as it was predicted that CPT1A would increase in the adipocyte co-cultured breast cancer cells. Increased CPT1A has been associated with upregulated β-oxidation and oncogenic effects in breast cancer cells (104, 150, 151). Studies in mice have shown an increase in CPT1A protein is associated with prevention of obesity (152-154). This has only just begun to be replicated in humans and it has been observed that different variants of CPT1 can influence adiposity levels (155, 156). CPT1A can be transcriptionally regulated (157), but this does not appear to be occurring. It is possible that there is CPT1A regulation occurring at a level that doesn’t change total protein level but rather protein availability; this is through malonyl-CoA inhibition. CPT1A is primarily regulated by concentration of malonyl-CoA (157). Malonyl-CoA competitively binds with high affinity to the CPT1A carnitine site, and it binds non-competitively with low affinity to the CPT1A palmitate site (158, 159). When malonyl-CoA binds CPT1A it causes reversible inhibition of CPT1A activity (105, 157, 158). If the concentration of malonyl-CoA is 69 altered between adipocyte co-culture and breast cancer cells grown alone this could be indicative of altered β-oxidation. Malonyl-CoA concentration was not directly measured in this project. Instead levels of ACC1 and phosphorylated ACC were measured. ACC catalyses the conversion of acetyl-CoA to malonyl-CoA, and when phosphorylated is inactive (95). There was no difference in ACC1 or phospho-ACC level between MCF7 and MDA-MB231 cells. Also, neither MCF7 nor MDA-MB-231 cells showed a change in the level of ACC1 between adipocyte co-cultured cells and cells grown alone. ACC1 level was not expected to change and so this result was in support of the hypothesis. It was expected that any changes in ACC would be seen as inhibition of ACC via phosphorylation rather than a transcriptional change. Phosphorylated ACC was significantly lower in cells grown alone compared to those grown in adipocyte co-culture for both MCF7 and MDAMB-231 cell lines. This is an important result and is supported by the literature on phosphorylated ACC function. When ACC is phosphorylated it is inhibited and cannot catalyse the acetyl-CoA to malonyl-CoA reaction; this leads to lower malonyl-CoA concentrations, increased availability of CPT1A, and increased β-oxidation (96, 102). This means that the adipocyte co-cultured cells have more available CPT1A which can be used to move fatty acids into the mitochondria for β-oxidation. It appears that the coculture of adipocytes with breast cancer cells is altering the cell phenotype so that the MCF7 and MDA-MB-231 cells can increase β-oxidation to take advantage of the fatty acids provided by CAA. One of the issues in ACC detection is that only ACC1 was probed for. It is thought that malonyl-CoA produced by ACC1 is primarily involved in the lipogenesis pathway, whilst malonyl-CoA from ACC2 is involved in β-oxidation (160, 161). ACC1 is found in the cytosol, but ACC2 is mostly located on the exterior of the mitochondria which is geographically advantageous for involvement in the β-oxidation pathway (162). Studies in mice have shown ACC2 knockouts have continuous β-oxidation (163). This finding is beginning to be replicated in humans with similar results (164). Both ACC1 and ACC2 possess the phosphorylation site detected by the anti-phospho-ACC antibody used in this study. Thus, both were taken into account in phosphorylated form but only ACC1 was detected in active form. 70 Through Western blotting, there have been changes observed in the β-oxidation pathway of MCF7 and MDA-MB-231 breast cancer cells during adipocyte co-culture. This is supported by previous work from Liu who detected increased β-oxidation in prostate cancer and even more closely by Nieman et al. (2011) who observed transfer of lipids to the mitochondrial matrix for β-oxidation in ovarian cancer cells co-cultured with adipocytes (17, 82). This is also supported by the wider research on breast cancer cell interactions with CAA. The increase in invasion and migration seen by Dirat et al. (2011) in breast cancer adipocyte co-culture could be partly due to an increase in β-oxidation (15). The glycerophosphate shuttle connects the cytosolic and mitochondrial metabolic pathways, allowing glycolytic NADH to contribute to oxidative metabolism (109). GPD2 is a regulatable component of this glycerophosphate shuttle (110). GPD2 was observed in MCF7 cells at 81 kDa and 68 kDa, whereas in MDA-MB-231 cells it was only observed at 81 kDa. There is no literature which describes the difference between the GPD2 isoforms or why this might be observed. Both cell lines had a decrease in GPD2 at 81kDa in adipocyte co-cultured cells compared to cells grown alone; this was observed as a non-significant trend in MDA-MB-231 cells. MCF7 cells did not show a change in total GPD2 level between cells co-cultured with adipocytes or grown alone. As mentioned above, the 81 kDa isoform was decreased in adipocyte co-cultured cells; the opposite trend was observed in the levels of the 68 kDa isoform. The 68 kDa isoform showed an increasing trend in adipocyte co-cultured cells compared to those grown alone. It appears that in MCF7 the two isoforms are differentially expressed during adipocyte co-culture. Recent literature has found GPD2 to be upregulated in glycolysis and in cancer, including breast cancer (111, 114). Due to the lack of literature around GPD2 isoforms it is difficult to interpret the findings any further. Thus far, it is seen that there were differences in GPD2 level in adipocyte co-cultured MCF7 cells. 4.3 Analysis of glycerol in conditioned media The final objective of this project was to measure adipocyte-derived glycerol and free fatty acids in adipocyte-breast tumour cell co-culture media. This would allow analysis of whether adipocytes have been induced to undergo lipolysis by co-culture with breast cancer cells. Due to resource and time constraints, only glycerol levels were measured in this project. It was hypothesised that glycerol levels would be lower in pre-adipocyte 71 samples compared to mature adipocyte, post co-culture, and 24 hour post co-culture samples. Also, that glycerol levels would be highest in post co-culture and 24 hour post co-culture samples. Lastly, it was hypothesised that glycerol level would not differ between cell lines for any of the conditioned media conditions. There were no significant changes in glycerol level observed across the different conditioned media samples between MCF7 and MDA-MB-231 cells. The other two hypotheses, however, were not supported. Glycerol levels tended to be lower, but were not significantly lower, in pre-adipocyte conditioned media compared to other conditioned media samples. Also, glycerol levels were not higher in post co-culture and 24 hour post co-culture conditioned media compared to other conditioned media samples. Levels of glycerol showed high variability between individual adipocyte samples and made comparisons difficult. As mentioned previously, glycerol is released during lipolysis in adipocytes (165); visual observations during co-culture showed loss of lipid content (Supplementary Figure 5.1). This lipid loss should therefore result in increased glycerol in the conditioned media. Glycerol was detected at a similar level in mature adipocyte and post co-culture conditioned media. If breast cancer cells are both causing increased glycerol release and consuming glycerol, this would be an expected result. In the mitochondrial stress test it was seen that MCF7 cells in adipocyte co-culture have upregulated glycolysis; it is suspected that this is to take advantage of the glycerol released by CAA. If the cells are utilising this glycerol this would account for the lack in glycerol level change between these conditions. The glycerol assay and mitochondrial stress test results support each other in a model of upregulated glycolysis. This interpretation is also supported by current Warburg theory and two-compartment tumour metabolism models (63, 80). It notes the co-cultured cells upregulating glycolysis, without diminishing oxidative respiration, and concedes that this is made possible by the release and subsequent uptake of glycerol from neighbouring adipocytes. 72 4.4 Future Work One of the key strengths in this study was the availability of a wide range of resources, knowledge, and previously optimised protocols. The combination of the Seahorse Extracellular Flux Analyser with the availability of human breast adipose samples and the adipocyte co-culture system allowed for a unique angle on metabolism in breast cancer cells. The main limit in the scope of this study, however, was the amount of work planned. After beginning optimisations it became clear that not all experiments would be completed due to time restrictions. This led to different sections of the study being completed to different extents and ended in the results not overlapping for interpretation between sections. There were no clear links between the results obtained from the Extracellular Flux Analyser and the results from Western blotting. The optimisation of procedures and the preliminary data gathered, nevertheless, has set up the capability to continue researching this area to complete the experiments originally planned. With further work, the Extracellular Flux Analyser can be used to observe βoxidation in MCF7 and MDA-MB-231 cells and measurements of free fatty acids can made in conditioned media. The results from this will be directly comparable with the preliminary observations of upregulated β-oxidation in adipocyte co-cultured MCF7 and MDA-MB-231 cells that was observed using Western blotting. Another limiting factor was the use of the Transwell co-culture. This system has no physical contact between cell populations and only allows for transfer of soluble factors between the adipocytes and the breast cancer cells. This means that all further analysis was based on the effects of secreted factors and therefore was not representative of physiological interactions in vivo. This allows for easier analysis post-co-culture and can elucidate the contribution of soluble factor secretions on cellular behaviour. It is unable to include potential changes due to cell-cell contact dependent communication. To get a more accurate interpretation of metabolic changes occurring in adipocyte cocultured MCF7 cells future work must include an appropriate control. When compared to a relevant control it may be apparent that the adipocytes have provided the co-cultured MCF7 cells with observable metabolic advantages. Only MCF7 cells were investigated in adipocyte co-culture, which gives a narrow view of metabolic changes in breast cancer cells. To expand the understanding of changes other cell lines should be examined. This 73 can begin with the MDA-MB-231 cell line which has already been metabolically characterised and provides a contrast to MCF7 and it is typically viewed as a more aggressive breast cancer cell line. The optimisation of the Seahorse fatty acid oxidation stress test has begun paving the way for investigation of this metabolic pathway on the Seahorse. This is the most important assay to finish optimising to assist in answering the hypothesis. This assay will allow further investigation to see if the signs of β-oxidation upregulation during adipocyte-breast cancer cell co-culture in MCF7 and MDA-MB-231 cells observed in Western blotting, are supported by more direct observation of metabolic alterations. This will help determine if adipocyte co-culture is upregulating β-oxidation in the breast cancer cells to take advantage of the fatty acids released by adipocytes. The observation that co-culture with adipocytes may be causing cancer cells to upregulate glycolysis should be further explored. This can be by adding the glycolysis stress test to the panel of Seahorse experiments that will be conducted in the future. This stress test provides a more detailed look at glycolysis and measures parameters of glycolysis, glycolytic capacity, glycolytic reserve, and non-glycolytic acidification (166). By undertaking further analysis of glycolysis it will be possible to provide a well-rounded view of metabolism during adipocyte-breast cancer cell co-culture. The results from Western blotting are promising and already show alterations in the βoxidation pathway during adipocyte co-culture with breast cancer cells. The changes shown so far in the β-oxidation pathway support the need for further investigation of this metabolic pathway during adipocyte-breast cancer cell co-culture. The proteins investigated already were chosen due to their key roles in the pathway. There are however other proteins in this pathway that can be observed to give a more detailed view of what is changing during adipocyte-breast cancer cell co-culture. AMPK is the primary cause of ACC phosphorylation, and as such, can be probed (96). Western blotting can also be carried out on malonyl-CoA, to directly measure levels of the key inhibitory protein in this β-oxidation pathway (97). ACC2 is a CPT1A inhibitor involved in the β-oxidation pathway (160, 161). It has been observed that ACC2 -/mice have continuous β-oxidation (163) and as such, ACC2 would be interesting to probe. CPT1C is a different CPT1 isoform that has been shown to be upregulated in cancer (167). This was also tested this in MCF7 cells and it was found that upregulation 74 of CPT1C caused upregulated β-oxidation and increased ATP synthesis. It is possible that although literature has focused on CPT1A there is a role for CPT1C in cancer as well. The remaining proteins in the CPT system can also be probed; this includes carnitine:acyl-carnitine translocase (CACT) and carnitine-palmitoyl transferase 2 (CPT2). CACT is located on the intermembrane side of the mitochondrial inner membrane and is responsible for moving fatty acids across the inner membrane into the matrix space (168). CPT2 is located on the matrix side of the inner membrane and prepares the fatty acids to be oxidised by removing carnitine (168). Both of these proteins do not appear to limit the rate of mitochondrial fatty acid uptake (157) but may still be altered in adipocyte co-culture if the β-oxidation pathway is upregulated. Lastly, long-chain acyl-CoA synthetase (LCAS) is another protein involved in B-oxidation. It prepares fatty acids in the cytosol for translocation by CPT1A (157). This is another important step that can be investigated. Although phosphorylated ACC was probed for, this antibody is not specific to ACC1 or ACC2 and can bind either. To elucidate which ACC isoform is being phosphorylated during co-culture with adipocytes, both ACC1 and ACC2 antibodies can be used to purify out these proteins by immunoprecipitation. These samples can then be probed with the anti-phospho-ACC antibody to determine which is phosphorylated, or if both are. Glycerol is only one of the metabolites that is released in lipolysis; another is fatty acids. Fatty acids were not measured in this project and are a key component to measure. This will allow linking of changes observed in Western blotting and Seahorse stress tests, with the visual observations of lipid loss which was made in co-culture. 4.5 Conclusion In this study two key observations were made; adipocyte co-cultured MCF7 cells showed upregulated glycolysis, and both co-cultured MCF7 and MDA-MB-231 cells displayed signs of upregulated β-oxidation. This study supports previous work that suggests adipocytes in the breast tumour microenvironment are capable of providing energetic advantage to metastasizing breast cancer cells. This has implications in the treatment, and outcomes, of overweight and obese breast cancer patients. If the mechanism(s) through which excess adiposity is having an adverse effect on patients can be determined, then this pathway can be used in development of targeted treatments. 75 5 Appendices 5.1 Media 5.1.1 Pre-adipocyte growth media Gibco® DMEM/F-12 medium [+] L-Glutamine [+] 2.438g/L sodium bicarbonate (Thermo Fisher Scientific) supplemented with 10% foetal bovine serum (FBS) and 1% Antibiotic Antimycotic Solution (100X) (Sigma-Aldrich). 5.1.2 Serum free media Gibco® DMEM/F-12 medium [+] L-Glutamine [+] 2.438g/L sodium bicarbonate supplemented with 1% Antibiotic Antimycotic Solution (100X). 5.1.3 Adipocyte differentiation media Made in serum free Gibco® DMEM/F-12 medium [+] L-Glutamine [+] 2.438g/L sodium bicarbonate supplemented with 1% Antibiotic Antimycotic Solution (100X). 0.5 mM 3-Isobutyl-1-methylxanthine (IBMX) 100 nM Insulin 100 nM Dexamethasone 2 nM 3,3′,5-triiodo-l-thyronine (T3) 1 µM Rosiglitazone 33 µM Biotin 17 µM Pantothenic acid 10 µg/mL Transferrin Alternative differentiation media –IBMX contains same final concentrations of above components minus IBMX. (Differentiation media is sterile filtered (Millex®-GV filter unit PVDF 0.22µm, Merck Millipore Ltd.) before use) 5.1.4 Adipocyte maintenance media Made in serum free Gibco® DMEM/F-12 medium [+] L-Glutamine [+] 2.438g/L sodium bicarbonate supplemented with 1% Antibiotic Antimycotic Solution (100X). 10 nM Insulin 10 nM Dexamethasone (Maintenance media is sterile filtered before use) 76 5.1.5 Breast cancer cell media Gibco® RPMI 1640 medium [+] L-Glutamine (Thermo Fisher Scientific) supplemented with 10% FBS and 1% Antibiotic Antimycotic Solution (100X). 5.1.6 MTT assay media Gibco® RPMI 1640 medium [+] L-Glutamine [-] Phenol Red (Thermo Fisher Scientific). 0.5 mg/ml 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) 5.1.7 Mitochondrial stress test assay media Made in XF assay medium (Seahorse Biosciences). 2 mM Pyruvate 0.4% Glucose Prepared fresh. Warmed to 37°C and adjusted to pH 7.35 using 0.1M sodium hydroxide. Sterile filtered before use. 5.1.8 Substrate limited media Made in Gibco® DMEM (-glutamine, -glucose, -HEPES) (Thermo Fisher Scientific). 0.5 mM Glucose 1 mM GlutaMAX 0.5 mM L-carnitine 1% FBS 5.1.9 FAO assay media (1X) Made in H2O. 20% FAO assay media 5X (Appendix 5.1.10) 2.5 mM Glucose 0.5 mM L-carnitine 5 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) Prepared fresh. Warmed to 37°C and adjusted to pH7.35 using 0.1M sodium hydroxide. Sterile filtered before use. 77 5.1.10 FAO assay media (5X) Made in distilled H2O. 555 mM Sodium chloride 23.5 mM Potassium chloride 6.25 mM Calcium chloride 10 mM Magnesium sulfate 6 mM Disodium phosphate 5.2 Buffers and solutions 5.2.1 Phosphate buffered saline (PBS) Made in distilled H2O. 1 PBS tablet/200ml (Sigma-Aldrich) Autoclaved and stored at room temperature. 5.2.2 Cell-Tak solution Made in sterile filtered pH8 100mM sodium bicarbonate solution. 22.4 µg/ml Cell-Tak™ (Corning) 5.2.3 Solubilisation solution Made in isopropanol. 10% Triton-X 1% Hydrochloric acid 5.2.4 Lysis buffer Made in RIPA (Appendix 5.2.5). 10% Phosphatase inhibitor (10X) (PhosSTOP, Roche) 2% Protease inhibitor (50X) (cOmplete tablets, Roche) 5.2.5 RIPA buffer Made in distilled H2O. 150 mM Sodium chloride 5% Tris 1% Octylphenoxypolyethoxyethanol (nonidet P-40) 0.5% w/v Sodium deoxycholate 0.1% Sodium dodecyl sulfate (SDS) 78 50% 5.2.6 Sample loading buffer LDS sample loading buffer (4x) (Thermo Fisher Scientific) 20% Dithiothreitol (DTT) (1M) 30% dH2O 5.2.7 Running buffer Made in distilled H2O. 5% 3-(N-morpholino) propanesulfonic acid (MOPS) SDS NuPage (20X) 5.2.8 Transfer buffer Made in distilled H2O. 10% Methanol 5% Bolt transfer buffer (20X) 5.2.9 TBS-T Made in distilled H2O. 10% 10X TBS 0.1% Tween-20 5.2.10 Mild stripping buffer Made in distilled H2O. 1.5% w/v Glycine 1% Tween-20 0.1% SDS Adjusted to pH2.2 using 12M hydrochloric acid. 79 Post co-culture adipocytes Mature adipocytes Pre-adipocytes 5.3 Supplementary Figures Figure 5.1: Representative photographs of adipocyte samples before and after co-culture with breast cancer cells. A) 20X AT047 pre-adipocyte. B) 10X AT062 and AT063 mature adipocytes, C) 10X AT062 and AT063 adipocytes post co-culture. 80 MDA-MB-231 1x104 2x104 1x105 5x105 MCF7 Figure 5.2: Representative photographs of cell seeding for MCF7 and MDA-MB-231 from 5x105-1x104 in an XF 24-well plate. Photographs are 10X from the centre of the well. 81 MCF7 MDA-MB-231 1x105 edge 1x105 centre A) 1x104 edge 1x104 centre B) Figure 5.3: Comparison of seeding at the centre and edge of XF 24-well plate for MCF7 and MDA-MB231 cells using 10X representative photographs. A) Seeding at 1x105 for MCF7 and MDA-MB-231. B) Seeding at 1x104 for MCF7 and MDA-MB-231. Top photographs from the centre of the well, bottom photographs from the edge of the well. 82 1 MCF7 0.9 MDA-MB-231 0.8 Absorbance 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 5 x 10⁵ 1 x 10⁵ 5 x 10⁴ 2 x 10⁴ 1 x 10⁴ 1 x 10³ 1 x 10² 0 Density (cells/well) Figure 5.4: MTT assay showing viability of MCF7 and MDA-MB-231 cells at varying densities. 83 AT053 25 25 Glycerol (µM) Glycerol (µM) 20 15 10 5 20 15 10 5 0 0 Mature adipocyte Mature adipocyte 100 50 150 100 50 0 0 Mature adipocyte Mature adipocyte Post co- 24hr post culture co-culture AT065 30 25 20 15 10 5 Post co- 24hr post culture co-culture AT062 80 Glycerol (µM) Glycerol (µM) Post co- 24hr post culture co-culture AT056 200 Glycerol (µM) Glycerol (µM) Post co- 24hr post culture co-culture AT055 150 60 40 20 0 0 Mature adipocyte Post co- 24hr post culture co-culture Mature adipocyte Post co- 24hr post culture co-culture AT063 50 Glycerol (µM) AT054 30 40 30 MCF7 20 MDA-MB-231 10 0 Mature adipocyte Post co- 24hr post culture co-culture Figure 5.5: Glycerol levels in varying stages of individual adipocyte co-cultures with MCF7 and MDAMB-231 cells. 84 A) B) MCF7 MDA-MB-231 Figure 5.6: MTT assays of MCF7 and MDA-MB-231 cells grown on inserts or not on inserts as controls to adipocyte co-cultured cells. A) Comparison of MCF7 controls grown on, or not on, inserts (p=0.119). B) Comparison of MDA-MB-231 controls grown on, or not on, inserts (p=0.634). 85 References 1. Organization WH. Obesity and overweight [Internet]: World Health Organization; 2015 [updated Jan 2015; cited 2015 May 26]. Available from: http://www.who.int/mediacentre/factsheets/fs311/en/. 2. Wang YC, McPherson K, Marsh T, Gortmaker SL, Brown M. Health and economic burden of the projected obesity trends in the USA and the UK. Lancet. 2011;378(9793):815-25. 3. Withrow D, Alter DA. The economic burden of obesity worldwide: a systematic review of the direct costs of obesity. Obes Rev. 2011;12(2):131-41. 4. Health Mo. The health of New Zealand adults 2011/12: Key findings of the New Zealand health survey. Wellington, New Zealand: 2012 Dec. 5. Health Mo. New Zealand health survey: Annual update of key findings 2012/13. Wellington, New Zealand: 2013. 6. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D. Global cancer statistics. CA Cancer J Clin. 2011;61(2):69-90. 7. Health Mo. Cancer: New registrations and deaths 2011. Wellington, New Zealand: 2014 Sep. 8. Health Mo. The price of cancer: The public price of registered cancer in New Zealand. Wellington, New Zealand: 2011 Sep. 9. Anderson GL, Neuhouser ML. Obesity and the risk for premenopausal and postmenopausal breast cancer. Cancer Prev Res (Phila). 2012;5(4):515-21. 10. Carmichael AR. Obesity and prognosis of breast cancer. Obes Rev. 2006;7(4):333-40. 11. Fusenig NE, Mueller MM. Friends or foes - bipolar effects of the tumour stroma in cancer. Nat Rev Cancer. 2004;4(11):839-49. 12. Joyce JA, Pollard JW. Microenvironmental regulation of metastasis. Nat Rev Cancer. 2009;9(4):239-52. 13. Elliott BE, Tam SP, Dexter D, Chen ZQ. Capacity of adipose tissue to promote growth and metastasis of a murine mammary carcinoma: effect of estrogen and progesterone. Int J Cancer. 1992;51:416-24. 14. Sundaram S, Johnson AR, Makowski L. Obesity, metabolism and the microenvironment: Links to cancer. J Carcinog. 2013;12:19. 15. Dirat B, Bochet L, Dabek M, Daviaud D, Dauvillier S, Majed B, et al. Cancerassociated adipocytes exhibit an activated phenotype and contribute to breast cancer invasion. Cancer Res. 2011;71(7):2455-65. 16. Bochet L, Meulle A, Imbert S, Salles B, Valet P, Muller C. Cancer-associated adipocytes promotes breast tumor radioresistance. Biochem Biophys Res Commun. 2011;411(1):102-6. 17. Nieman KM, Kenny HA, Penicka CV, Ladanyi A, Buell-Gutbrod R, Zillhardt MR, et al. Adipocytes promote ovarian cancer metastasis and provide energy for rapid tumour growth. Nat Med. 2011;17(11):1498-503. 18. Pavlides S, Tsirigos A, Vera I, Flomenberg N, Frank PG, Casimiro MC, et al. Transcriptional evidence for the "reverse warburg effect" in human breast cancer tumor stroma and metastasis: similarities with oxidative stress, inflammation, alzheimer's disease, and "neuron-glia metabolic coupling". Aging. 2010;2(4):185-99. 86 19. Salem AF, Whitaker-Menezes D, Lin Z, Martinez-Outschoorn UE, Tanowitz HB, Al-Zoubi MS, et al. Two-compartment tumor metabolism: autophagy in the tumor microenvironment and oxidative mitochondrial metabolism (OXPHOS) in cancer cells. Cell Cycle. 2012;11(13):2545-56. 20. Tchernof A, Després J-P. Pathophysiology of human visceral obesity: an update. Physiol Rev. 2013;93:359-404. 21. Organization WH. Global health risks: mortality and burden of disease attributable to selected major risks. Geneva, Switzerland: 2009. 22. Health Mo. A portrait of health: Key results of the 2006/07 New Zealand health survey. Wellington, New Zealand: 2008 Jun. 23. Health Mo. Actions to encourage healthy weight and healthy lifestyles. Wellington, New Zealand: 2014 May. 24. Gonzalez ABd, Hartge P, Hartge P, Flint AJ, Hannan L, MacInnis RJ, et al. BodyMass Index and mortality among 1.46 million white adults. New Engl J Med. 2010;363(23):2211-9. 25. Ogden CL, Carroll MD, Kit BK, Flegal KM. Prevalence of childhood and adult obesity in the United States, 2011-2012. JAMA. 2014;311(8):806-14. 26. Calle EE, Kaaks R. Overweight, obesity and cancer: epidemiological evidence and proposed mechanisms. Nat Rev Cancer. 2004;4(8):579-91. 27. Renehan AG, Tyson M, Egger M, Heller RF, Zwahlen M. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet. 2008;371(9612):569-78. 28. Wolin KY, Carson K, Colditz GA. Obesity and cancer. Oncologist. 2010;15(6):556-65. 29. Goodwin PJ, Stambolic V. Impact of the obesity epidemic on cancer. Annu Rev Med. 2015;66:281-96. 30. Iyengar NM, Hudis CA, Dannenberg AJ. Obesity and cancer: local and systemic mechanisms. Annu Rev Med. 2015;66:297-309. 31. Hursting SD, Dunlap SM. Obesity, metabolic dysregulation, and cancer: a growing concern and an inflammatory (and microenvironmental) issue. Ann N Y Acad Sci. 2012;1271:82-7. 32. Roberts DL, Dive C, Renehan AG. Biological mechanisms linking obesity and cancer risk: new perspectives. Annu Rev Med. 2010;61:301-16. 33. Vucenik I, Stains JP. Obesity and cancer risk: evidence, mechanisms, and recommendations. Ann N Y Acad Sci. 2012;1271:37-43. 34. Foundation NZBC. What is breast cancer [Internet]: New Zealand Breast Cancer Foundation; 2013 [cited 2015 May 25]. Available from: http://www.nzbcf.org.nz/BREASTCANCER/Overview/WhatisBreastCancer.aspx. 35. Prat A, Perou CM. Deconstructing the molecular portraits of breast cancer. Mol Oncol. 2011;5(1):5-23. 36. Ban KA, Godellas CV. Epidemiology of breast cancer. Surg Oncol Clin N Am. 2014;23(3):409-22. 37. Abe R, Kumagai N, Kimura M, Hirosaki A, Nakamura T. Biological characteristics of breast cancer in obesity. Tohoku J Exp Med. 1976;120:351-59. 38. Ligibel J. Obesity and breast cancer. Oncology. 2011;25(11):994-1000. 87 39. Ewertz M, Jensen MB, Gunnarsdottir KA, Hojris I, Jakobsen EH, Nielsen D, et al. Effect of obesity on prognosis after early-stage breast cancer. J Clin Oncol. 2011;29(1):25-31. 40. Rose DP, Vona-Davis L. Biochemical and molecular mechanisms for the association between obesity, chronic inflammation, and breast cancer. BioFactors. 2014;40(1):1-12. 41. Majed B, Moreau T, Senouci K, Salmon RJ, Fourquet A, Asselain B. Is obesity an independent prognosis factor in woman breast cancer? Breast Cancer Res Treat. 2008;111(2):329-42. 42. Protani M, Coory M, Martin JH. Effect of obesity on survival of women with breast cancer: systematic review and meta-analysis. Breast Cancer Res Treat. 2010;123:627-35. 43. Rose DP, Vona-Davis L. Influence of obesity on breast cancer receptor status and prognosis. Expert Rev Anticancer Ther. 2009;9(8):1091-101. 44. Lorincz AM, Sukumar S. Molecular links between obesity and breast cancer. Endocr Relat Cancer. 2006;13(2):279-92. 45. Dieudonne M-N, Machinal-Quelin F, Serazin-Leroy V, Leneveu M-C, Pecquery R, Giudicelli Y. Leptin mediates a proliferative response in human MCF7 breast cancer cells. Biochem Biophys Res Commun. 2002;293:622-28. 46. Pavlides S, Vera I, Gandara R, Sneddon S, Pestell RG, Mercier I, et al. Warburg meets autophagy: cancer-associated fibroblasts accelerate tumor growth and metastasis via oxidative stress, mitophagy, and aerobic glycolysis. Antioxid Redox Signal. 2012;16(11):1264-84. 47. Polyak K, Kalluri R. The role of the microenvironment in mammary gland development and cancer. Cold Spring Harb Perspect Biol. 2010;2(11):1-14. 48. Dvorak HF. Tumours: wounds that do not heal. N Engl J Med. 1986;315(26):1650-9. 49. Paget S. The distribution of secondary growths in cancer of the breast. Lancet. 1889;133(3421):571-3. 50. Fidler IJ, Poste G. The “seed and soil” hypothesis revisited. Lancet Oncol. 2008;9(8):808. 51. Bissell MJ, Hall HG, Parry G. How does the extracellular matrix direct gene expression? J Theor Biol. 1982;99:31-68. 52. Bissell MJ, Ram TG. Regulation of functional cytodifferentiation and histogenesis in mammary epithelial cells: role of the extracellular matrix. Environ Health Perspect. 1989;80:61-7. 53. Artacho-Cordon A, Artacho-Cordon F, Rios-Arrabal S, Calvente I, Nunez MI. Tumor microenvironment and breast cancer progression: a complex scenario. Cancer Biol Ther. 2012;13(1):14-24. 54. Ronnov-Jessen L, Bissell MJ. Breast cancer by proxy: can the microenvironment be both the cause and consequence? Trends Mol Med. 2009;15(1):5-13. 55. Park J, Morley TS, Kim M, Clegg DJ, Scherer PE. Obesity and cancer-mechanisms underlying tumour progression and recurrence. Nat Rev Endocrinol. 2014;10(8):455-65. 56. Wiseman BS, Werb Z. Stromal effects on mammry gland development and breast cancer. Science. 2002;296:1046-9. 88 57. DeCosse JJ, Gossens CL, Kuzma JF. Breast cancer: induction of differentiation by embryonic tissue. Science. 1973;181:1057-8. 58. Place AE, Huh SJ, Polyak K. The microenvironment in breast cancer progression: biology and implications for treatment. Breast Cancer Res. 2011;13(227):1-11. 59. Roskelley CD, Bissell MJ. The dominance of the microenvironment in breast and ovarian cancer. Semin Cancer Biol. 2002;12(2):97-104. 60. Lin EY, Pollard JW. Tumor-associated macrophages press the angiogenic switch in breast cancer. Cancer Res. 2007;67(11):5064-6. 61. Mao Y, Keller ET, Garfield DH, Shen K, Wang J. Stromal cells in tumor microenvironment and breast cancer. Cancer Metastasis Rev. 2013;32(1-2):303-15. 62. Karnoub AE, Dash AB, Vo AP, Sullivan A, Brooks MW, Bell GW, et al. Mesenchymal stem cells within tumour stroma promote breast cancer metastasis. Nature. 2007;449(7162):557-63. 63. Pavlides S, Whitaker-Menezes D, Castello-Cros R, Flomenberg N, Witkiewicz AK, Frank PG, et al. The reverse Warburg effect: Aerobic glycolysis in cancer associated fibroblasts and the tumor stroma. Cell Cycle. 2009;8(23):3984-4001. 64. Hasebe T, Sasaki S, Imoto S, Ochiai A. Highly proliferative fibroblasts forming fibrotic focus govern meastasis of invasive ductal carcinoma of the breast. Mod Pathol. 2001;14(4):325-37. 65. Hu M, Peluffo G, Chen H, Gelman R, Schnitt S, Polyak K. Role of COX-2 in epithelial-stromal cell interactions and progression of ductal carcinoma in situ of the breast. Proc Natl Acad Sci U S A. 2009;106(9):3372-7. 66. Shekhar MPV, Werdell J, Santner SJ, Pauley RJ, Tait L. Breast stroma plays a dominant regulatory role in breast epithelial growth and differentiation: implications for tumour development and progression. Cancer Res. 2008;61:1320-6. 67. Ouchi N, Parker JL, Lugus JJ, Walsh K. Adipokines in inflammation and metabolic disease. Nat Rev. 2011;11:85-97. 68. Carter JC, Church FC. Mature breast adipocytes promote breast cancer cell motility. Exp Mol Pathol. 2012;92(3):312-7. 69. Wang YY, Lehuede C, Laurent V, Dirat B, Dauvillier S, Bochet L, et al. Adipose tissue and breast epithelial cells: a dangerous dynamic duo in breast cancer. Cancer Lett. 2012;324(2):142-51. 70. Manabe Y, Toda S, Miyazaki K, Sugihara H. Mature adipocytes, but not preadipocytes, promote the growth of breast carcinoma cells in collagen gel matrix culture through cancer-stromal cell interactions. J Pathol. 2003;201(2):221-8. 71. Lee Y, Jung WH, Koo JS. Adipocytes can induce epithelial-mesenchymal transition in breast cancer cells. Breast Cancer Res Treat. 2015;153(2):323-35. 72. Motrescu ER, Rio MC. Cancer cells, adipocytes and matrix metalloproteinase 11: a vicious tumor progression cycle. Biol Chem. 2008;389(8):1037-41. 73. Wang YY, Lehuede C, Laurent V, Dirat B, Dauvillier S, Bochet L, et al. Adipose tissue and breast epithelial cells: a dangerous dynamic duo in breast cancer. Cancer Lett. 2012;324(2):142-51. Figure 3, Invading tumor cells are able to modify adipocytes phenotype, which, in turn, stimulate cancer cells aggressive behavior; p. 6. 74. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646-74. 89 75. Seyfried TN, Flores RE, Poff AM, D'Agostino DP. Cancer as a metabolic disease: implications for novel therapeutics. Carcinogenesis. 2014;35(3):515-27. 76. Cairns RA, Harris IS, Mak TW. Regulation of cancer cell metabolism. Nat Rev Cancer. 2011;11(2):85-95. 77. Icard P, Kafara P, Steyaert JM, Schwartz L, Lincet H. The metabolic cooperation between cells in solid cancer tumors. Biochim Biophys Acta. 2014;1846(1):216-25. 78. Warburg O. The prime cause and prevention of cancer. Nobel-Laureates; 1969; Lindau, Lake Constance, Germany: Konrad Triltsch, Würzburg, Germany. 79. Warburg O, Wind F, Negelein E. The metabolism of tumours in the body. J Gen Physiol. 1927;8(6):519-30. 80. Chen X, Qian Y, Wu S. The Warburg effect: evolving interpretations of an established concept. Free Radic Biol Med. 2015;79:253-63. 81. Koppenol WH, Bounds PL, Dang CV. Otto Warburg's contributions to current concepts of cancer metabolism. Nat Rev Cancer. 2011;11(5):325-37. 82. Liu Y. Fatty acid oxidation is a dominant bioenergetic pathway in prostate cancer. Prostate Cancer Prostatic Dis. 2006;9(3):230-4. 83. Baehrecke EH. Autophagy: dual roles in life and death. Nat Rev. 2005;6:505-10. 84. Codogno P, Meijer AJ. Autophagy and signaling: their role in cell survival and cell death. Cell Death Differ. 2005;12 Suppl 2:1509-18. 85. Carracedo A, Cantley LC, Pandolfi PP. Cancer metabolism: fatty acid oxidation in the limelight. Nat Rev Cancer. 2013;13(4):227-32. 86. Huynh FK, Green MF, Koves TR, Hirschey MD. Measurement of fatty acid oxidation rates in animal tissues and cell lines. Methods Enzymol. 2014;542:391-405. 87. Waris G, Ahsan H. Reactive oxygen species: role in the development of cancer and various chronic conditions. J Carcinog. 2006;5:14. 88. Liou GY, Storz P. Reactive oxygen species in cancer. Free Radic Res. 2010;44(5):479-96. 89. Zaidi N, Lupien L, Kuemmerle NB, Kinlaw WB, Swinnen JV, Smans K. Lipogenesis and lipolysis: the pathways exploited by the cancer cells to acquire fatty acids. Prog Lipid Res. 2013;52(4):585-9. 90. Kuhajda FP, Pizer ES, Li JN, Mani NS, Frehywot GL, Townsend CA. Synthesis and antitumor activity of an inhibitor of fatty acid synthase. Proc Natl Acad Sci U S A. 2000;97(7):3450-4. 91. Menendez JA, Lupu R. Fatty acid synthase and the lipogenic phenotype in cancer pathogenesis. Nat Rev Cancer. 2007;7(10):763-77. 92. Hopperton KE, Duncan RE, Bazinet RP, Archer MC. Fatty acid synthase plays a role in cancer metabolism beyond providing fatty acids for phospholipid synthesis or sustaining elevations in glycolytic activity. Exp Cell Res. 2014;320(2):302-10. 93. Louie SM, Roberts LS, Mulvihill MM, Luo K, Nomura DK. Cancer cells incorporate and remodel exogenous palmitate into structural and oncogenic signaling lipids. Biochim Biophys Acta. 2013;1831(10):1566-72. 94. Nieman KM, Kenny HA, Penicka CV, Ladanyi A, Buell-Gutbrod R, Zillhardt MR, et al. Adipocytes promote ovarian cancer metastasis and provide energy for rapid tumour growth. Nat Med. 2011;17(11):1498-503. Figure 4(g), FABP4 has a key role in the interaction of cancer cells and adipocytes; p. 502. 90 95. Witters LA, Widmer J, King AN, Fassihi K, Kuhajda F. Identification of human acetyl-CoA carboxylase isozymes in tissue and in breast cancer cells. Int J Biochem. 1994;26(4):589-94. 96. Yamauchi T, Kamon J, Minokoshi Y, Ito Y, Waki H, Uchida S, et al. Adiponectin stimulates glucose utilization and fatty-acid oxidation by activating AMP-activated protein kinase. Nat Med. 2002;8(11):1288-95. 97. McGarry JD, Mannaerts GP, Foster DW. A possible role for malonyl-CoA in the regulation of hepatic fatty acid oxidation and ketogenesis. J Clin Invest. 1977;60:26570. 98. Tong L, Harwood HJ, Jr. Acetyl-coenzyme A carboxylases: versatile targets for drug discovery. J Cell Biochem. 2006;99(6):1476-88. 99. Carling D. The AMP-activated protein kinase cascade--a unifying system for energy control. Trends Biochem Sci. 2004;29(1):18-24. 100. Fogarty S, Hardie DG. Development of protein kinase activators: AMPK as a target in metabolic disorders and cancer. Biochim Biophys Acta. 2010;1804(3):581-91. 101. Jeon SM, Hay N. The double-edged sword of AMPK signaling in cancer and its therapeutic implications. Arch Pharm Res. 2015;38(3):346-57. 102. Minokoshi Y, Kim Y-B, Peroni OD, Fryer LGD, Muller C, Carling D, et al. Leptin stimulates fatty-acid oxidation by activating AMP-activated protein kinase. Nature. 2002;415:339-43. 103. Janovska A, Hatzinikolas G, Staikopoulos V, McInerney J, Mano M, Wittert GA. AMPK and ACC phosphorylation: effect of leptin, muscle fibre type and obesity. Mol Cell Endocrinol. 2008;284(1-2):1-10. 104. Linher-Melville K, Zantinge S, Sanli T, Gerstein H, Tsakiridis T, Singh G. Establishing a relationship between prolactin and altered fatty acid β-oxidation via carnitine palmitoyl transferase 1 in breast cancer cells. BMC Cancer. 2011;11(56). 105. McGarry JD, Brown NF. The mitochondrial carnitine palmitoyltransferase system. Eur J Biochem. 1997;244:1-14. 106. Murthy MSR, Pande SV. Malonyl-CoA binding site and the overt carnitine palmitoyltransferase activity reside on the opposite sides of the outer mitochondrial membrane. Proc Natl Acad Sci U S A. 1987;84:378-82. 107. Garrib A, McMurray WC. Purification and characterisation of glycerol-3phosphate dehydrogenase (flavin-linked) from rat liver mitochondria. J Biol Chem. 1986;261(17):8042-48. 108. UniProt. UniProtKB - P43304 (GPDM_HUMAN) [Internet]. [cited 2015 Sep 29]. Available from http://www.uniprot.org/uniprot/P43304#sequences. 109. Madiraju AK, Erion DM, Rahimi Y, Zhang XM, Braddock DT, Albright RA, et al. Metformin suppresses gluconeogenesis by inhibiting mitochondrial glycerophosphate dehydrogenase. Nature. 2014;510(7506):542-6. 110. Mracek T, Drahota Z, Houstek J. The function and the role of the mitochondrial glycerol-3-phosphate dehydrogenase in mammalian tissues. Biochim Biophys Acta. 2013;1827(3):401-10. 111. Chowdhury SK, Gemin A, Singh G. High activity of mitochondrial glycerophosphate dehydrogenase and glycerophosphate-dependent ROS production in prostate cancer cell lines. Biochem Biophys Res Commun. 2005;333(4):1139-45. 91 112. Miwa S, Brand MD. The topology of superoxide production by complex III and glycerol 3-phosphate dehydrogenase in Drosophila mitochondria. Biochim Biophys Acta. 2005;1709(3):214-9. 113. Mracek T, Holzerova E, Drahota Z, Kovarova N, Vrbacky M, Jesina P, et al. ROS generation and multiple forms of mammalian mitochondrial glycerol-3-phosphate dehydrogenase. Biochim Biophys Acta. 2014;1837(1):98-111. 114. MacDonald MJ, Warner TF, Mertz RJ. High activity of mitochondrial glycerol phosphate dehydrogenase in insulinomas and carcinoid and other tumors of the amine precursor uptake decarboxylation system. Cancer Res. 1990;50:7203-5. 115. Lee MJ, Wu Y, Fried SK. A modified protocol to maximize differentiation of human preadipocytes and improve metabolic phenotypes. Obesity (Silver Spring). 2005;20(12):2334-40. 116. Holliday DL, Speirs V. Choosing the right cell line for breast cancer research. Breast Cancer Research. 2011;13(215):1-7. 117. Bauer KR, Brown M, Cress RD, Parise CA, Caggiano V. Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2negative invasive breast cancer, the so-called triple-negative phenotype: a populationbased study from the California cancer Registry. Cancer. 2007;109(9):1721-8. 118. Griffiths CL, Olin JL. Triple negative breast cancer: a brief review of its characteristics and treatment options. J Pharm Pract. 2012;25(3):319-23. 119. Neve RM, Chin K, Fridlyand J, Yeh J, Baehner FL, Fevr T, et al. A collection of breast cancer cell lines for the study of functionally distinct cancer subtypes. Cancer Cell. 2006;10(6):515-27. 120. XF24 extracellular flux analyzer and prep station installation and operation manual. 1st ed. Revision 12/21/10. N. Billerica: Seahorse Bioscience Inc.; 2010. 121. User manual: XF cell mito stress test kit. U.S.A.: Seahorse Biosciences; 2010. 122. Bansal S, Leu AN, Gonzalez FJ, Guengerich FP, Chowdhury AR, Anandatheerthavarada HK, et al. Mitochondrial targeting of cytochrome P450 (CYP) 1B1 and its role in polycyclic aromatic hydrocarbon-induced mitochondrial dysfunction. J Biol Chem. 2014;289(14):9936-51. 123. Caicedo A, Fritz V, Brondello JM, Ayala M, Dennemont I, Abdellaoui N, et al. MitoCeption as a new tool to assess the effects of mesenchymal stem/stromal cell mitochondria on cancer cell metabolism and function. Sci Rep. 2015;5:9073. 124. Diers AR, Broniowska KA, Chang CF, Hill RB, Hogg N. S-Nitrosation of monocarboxylate transporter 1: inhibition of pyruvate-fueled respiration and proliferation of breast cancer cells. Free Radic Biol Med. 2014;69:229-38. 125. Kumar S, Donti TR, Agnihotri N, Mehta K. Transglutaminase 2 reprogramming of glucose metabolism in mammary epithelial cells via activation of inflammatory signaling pathways. Int J Cancer. 2014;134(12):2798-807. 126. Ozden O, Park SH, Wagner BA, Yong Song H, Zhu Y, Vassilopoulos A, et al. SIRT3 deacetylates and increases pyruvate dehydrogenase activity in cancer cells. Free Radic Biol Med. 2014;76:163-72. 127. Qiu F, Chen Y-R, Liu X, Chu C-Y, Shen L-J, Xu J, et al. Arginine starvation impairs mitochondrial respiratory function in ASS1-deficient breast cancer cells. Science Signaling. 2014;7(319):1-17. 128. Vaughan RA, Gannon NP, Garcia-Smith R, Licon-Munoz Y, Barberena MA, Bisoffi M, et al. β-alanine suppresses malignant breast epithelial cell aggressiveness 92 through alterations in metabolism and cellular acidity in vitro. Molecular Cancer. 2014;13(14). 129. Zhou T, Zhang B, Wei P, Du Y, Zhou H, Yu M, et al. Energy metabolism analysis reveals the mechanism of inhibition of breast cancer cell metastasis by PEG-modified graphene oxide nanosheets. Biomaterials. 2014;35(37):9833-43. 130. Immobilization of non-adherent cells with Cell-Tak for assay on the XF analyzer [Internet]. Seahorse Biosciences [cited 2015 Oct 10]. Available from http://www.seahorsebio.com/resources/tech-writing/protocol-non-adherent.pdf. 131. Simultaneously measuring oxidation of exogenous and endogenous fatty acids using the XF palmitate-BSA FAO substrate with the XF cell mito stress test [Internet]. Seahorse Biosciences [cited 2015 Sep 21]. Available from http://www.seahorsebio.com/resources/techwriting/XF_FAO_Assay_Technical_Brief.pdf. 132. Preparation of bovine serum albumin (BSA)-conjugated palmitate [Internet]. Seahorse Biosciences [cited 2015 May 19]. 133. Characterizing your cells: using OCR values to determine optimal seeding density [Internet]. Seahorse Biosciences [cited 2015 Oct 15]. Available from http://www.seahorsebio.com/resources/tech-writing/Tech-Hint-Basal-OCR.pdf. 134. Sun S, Han Y, Liu J, Fang Y, Tian Y, Zhou J, et al. Trichostatin A targets the mitochondrial respiratory chain, increasing mitochondrial reactive oxygen species production to trigger apoptosis in human breast cancer cells. PLoS One. 2014;9(3):e91610. 135. Zhao L, Zhao X, Zhao K, Wei P, Fang Y, Zhang F, et al. The alpha-tocopherol derivative ESeroS-GS induces cell death and inhibits cell motility of breast cancer cells through the regulation of energy metabolism. Eur J Pharmacol. 2014;745:98-107. 136. XFp cell mito stress test kit: user guide [Internet]. Seahorse Biosciences [cited 2015 Oct 21]. Available from http://www.seahorsebio.com/resources/pdfs/XFp_Cell_Mito_Stress_Test_Kit_User_G uide.pdf. 137. New cell characterization [Internet]. Seahorse Biosciences [cited 2015 Oct 22]. Available from http://seahorsebio.com/resources/basicprocedures/New_Cell_Line_Characterization.pdf. 138. XF stress test report generator: user guide [Internet]. Seahorse Biosciences [cited 2015 Oct 21]. Available from http://www.seahorsebio.com/support/software/PDF/XF_Stress_Test_Report_Generat or_User_Guide.pdf. 139. XF cell energy phenotype report generator: user guide [Internet]. Seahorse Biosciences [cited 2015 Oct 19]. Available from http://seahorsebio.com/support/software/PDF/XF_Cell_Energy_Phenotype_Test_Rep ort_Generator_User_Guide.pdf. 140. Seahorse XFe extracellular flux analyzers: the world's most advanced metabolic analyzers [Internet]. Seahorse Biosciences [cited 2015 Oct 15]. Available from http://www.seahorsebio.com/resources/pdfs/marketing/XFe_Brochure_LowRes.pdf. 141. de Ruijter TC, Veeck J, de Hoon JP, van Engeland M, Tjan-Heijnen VC. Characteristics of triple-negative breast cancer. J Cancer Res Clin Oncol. 2011;137(2):183-92. 93 142. Cooper GM. The cell: a molecular approach. 2nd ed. Sunderland: Sinauer Associates; 2000. 143. Brookes PS. Mitochondrial H(+) leak and ROS generation: an odd couple. Free Radic Biol Med. 2005;38(1):12-23. 144. Pelicano H, Zhang W, Liu J, Hammoudi N, Dai J, Xu R-H, et al. Mitochondrial dysfunction in some triple-negative breast cancer cell lines: role of mTOR pathway and therapeutic potential. Breast Cancer Research. 2014;16(434):1-16. 145. Gupta C, Tikoo K. High glucose and insulin differentially modulates proliferation in MCF-7 and MDA-MB-231 cells. J Mol Endocrinol. 2013;51(1):119-29. 146. Jeso V, Yang C, Cameron MD, Cleveland JL, Micalizio GC. Synthesis and SAR of Lehualide B: a marine-derived natural product with potent anti-multiple myeloma activity. ACS Chem Biol. 2013;8(6):1241-52. 147. Sohn Y-S, Tamir S, Song L, Michaeli D, Matouk I, Conlan AR, et al. NAF-1 and mitoNEET are central to human breast cancer proliferation by maintaining mitochondrial homeostasis and promoting tumor growth. PNAS. 2013;110(36). 148. Stoker HS. General, organic, & biological chemistry. 7th ed. Boston: Cengage Learning; 2014. p.949. 149. Kim S, Lee Y, Koo JS. Differential expression of lipid metabolism-related proteins in different breast cancer subtypes. PLoS One. 2015;10(3):e0119473. 150. Jambor de Sousa UL, Koss MD, Fillies M, Gahl A, Scheeder MR, Cardoso MC, et al. CPT1alpha over-expression increases long-chain fatty acid oxidation and reduces cell viability with incremental palmitic acid concentration in 293T cells. Biochem Biophys Res Commun. 2005;338(2):757-61. 151. Rubi B, Antinozzi PA, Herrero L, Ishihara H, Asins G, Serra D, et al. Adenovirusmediated overexpression of liver carnitine palmitoyltransferase I in INS1E cells : effects on cell metabolism and insulin secretion. Biochem J. 2002;364:219-26. 152. Fak F, Backhed F. Lactobacillus reuteri prevents diet-induced obesity, but not atherosclerosis, in a strain dependent fashion in Apoe-/- mice. PLoS One. 2012;7(10):e46837. 153. Orellana-Gavalda JM, Herrero L, Malandrino MI, Paneda A, Sol Rodriguez-Pena M, Petry H, et al. Molecular therapy for obesity and diabetes based on a long-term increase in hepatic fatty-acid oxidation. Hepatology. 2011;53(3):821-32. 154. Shimoda H, Hitoe S, Nakamura S, Matsuda H. Purple tea and its extracts suppress diet-induced fat accumulation in mice and human subjects by inhibiting fat absorption and enhancing hepatic carnitine palmitoyltransferase expression. Int J Biomed Sci. 2015;11(2):67-75. 155. Lemas DJ, Wiener HW, O'Brien DM, Hopkins S, Stanhope KL, Havel PJ, et al. Genetic polymorphisms in carnitine palmitoyltransferase 1A gene are associated with variation in body composition and fasting lipid traits in Yup'ik Eskimos. J Lipid Res. 2012;53(1):175-84. 156. Robitaille J, Houde A, Lemieux S, Perusse L, Gaudet D, Vohl MC. Variants within the muscle and liver isoforms of the carnitine palmitoyltransferase I (CPT1) gene interact with fat intake to modulate indices of obesity in French-Canadians. J Mol Med (Berl). 2007;85(2):129-37. 157. Kerner J, Hoppel C. Fatty acid import into mitochondria. Biochimica et Biophysica Acta. 2000;1486:1-17. 94 158. Lopez-Vinas E, Bentebibel A, Gurunathan C, Morillas M, de Arriaga D, Serra D, et al. Definition by functional and structural analysis of two malonyl-CoA sites in carnitine palmitoyltransferase 1A. J Biol Chem. 2007;282(25):18212-24. 159. Zammit VA. Carnitine palmitoyltransferase 1: central to cell function. IUBMB Life. 2008;60(5):347-54. 160. Abu-Elheiga L, Almarza-Ortega DB, Baldini A, Wakil SJ. Human acetyl-CoA carboxylase 2. J Biol Chem. 1997;272(16):10669-77. 161. Schreurs M, Kuipers F, van der Leij FR. Regulatory enzymes of mitochondrial beta-oxidation as targets for treatment of the metabolic syndrome. Obes Rev. 2010;11(5):380-8. 162. Abu-Elheiga L, Brinkley WR, Zhong L, Chirala SS, Woldegiorgis G, Wakil SJ. The subcellular localization of acetyl-CoA carboxylase 2. PNAS. 2000;97(4):1444-9. 163. Abu-Elheiga L, Matzuk MM, Abo-Hashema KAH, Wakil SJ. Continuous fatty acid oxidation and reduced fat storage in mice lacking acetyl-CoA carboxylase 2. Science. 2001;291(5513):2613-6. 164. Kreuz S, Schoelch C, Thomas L, Rist W, Rippmann JF, Neubauer H. Acetyl-CoA carboxylases 1 and 2 show distinct expression patterns in rats and humans and alterations in obesity and diabetes. Diabetes Metab Res Rev. 2009;25(6):577-86. 165. Diedrich J, Gusky HC, Podgorski I. Adipose tissue dysfunction and its effects on tumor metabolism. Horm Mol Biol Clin Investig. 2015;21(1):17-41. 166. XF glycolysis stress test kit: user guide. U.S.A.: Seahorse Biosciences; 2012. 167. Zaugg K, Yao Y, Reilly PT, Kannan K, Kiarash R, Mason J, et al. Carnitine palmitoyltransferase 1C promotes cell survival and tumor growth under conditions of metabolic stress. Genes Dev. 2011;25(10):1041-51. 168. Prip-Buus C, Thuillier L, Abadi N, Prasad C, Dilling L, Klasing J, et al. Molecular and enzymatic characterization of a unique carnitine palmitoyltransferase 1A mutation in the Hutterite community. Mol Genet Metab. 2001;73(1):46-54. 95